Public report — omi, published 5 Aug 2026.
Concrete security findings (CVE IDs, secret matches, dependency versions) are hidden in this version;
ask the repo owner for the full report.
587findings with an exact file:lineof 613 — the remainder are repo-wide signals (a dimension-level measurement, not a single line); open any file:line and verify
58/115dimensions across the health lenses1004 LoC · 3 projects — wide & deep
Executive summary
Read through the Production lens — the standard calibration. *Green* means good enough to run in production. The score is absolute and comparable across repos.
BasedHardware/omi is sound in substance but carries real gaps (52%). It is not in crisis, but the issues below raise the cost of changing it — friction its consumers ultimately inherit.
Most urgent: a critical security exposure was detected (see the Security & Compliance lens). Treat it as a priority regardless of the overall grade.
The area that most needs attention is Readiness (47%) — releases are harder to depend on — versioning, release notes and dependency hygiene are thin, so consumers can't easily tell what changed or trust an upgrade. Accessibility (49%) is the next concern — it raises ongoing delivery and operational cost.
Leadership focus, highest impact first: unused dependencies, declare unlisted imports explicitly,… (Dependency Hygiene); 8 Leaked secret finding(s) (Secret Scanning); tests that import the unreached modules (directly or through… (Test Coverage).
For scale: Hobby (~1,004 production lines); rebuilding it from scratch would take roughly ~3.8 person-years (~2–8 engineers). Approximate, ±~30%.
How the score is built — each lens's share of the headlineWidth is the lens's weight in the worst-heaviest fold (the weakest area pulls hardest); colour is that lens's own band. A lens fixes the score in proportion to its width.
~12407 LoC unreachable (115 file(s)) — that slice of this estimate buys code with zero runtime value; deleting it is the cheapest win in this report (the R7 card lists every file)
This codebase represents roughly ~3.8 person-years of build effort (about ~€550,000 to rebuild). Its weakest lens is Readiness at 47% — the part of that asset most exposed by the findings below.
How we model this: boilerplate at a scaffolding rate + logic × domain Standard (×1.3) — domain model, high decision density × a 0.7× quality factor, at €60–95/h; indicative, ±~30% · size measured directly from source. Indicative only — most sensitive to the hourly rate and the domain tier (both tunable in config).
Top priorities
The highest-leverage moves; the full ranked list is in the Roadmap below.
1
Resolve the 8 Leaked secret finding(s) in Secret Scanning — start with run.sh (2), oauth.py (2), public-build-values.json.
Of everything flagged, the best return on effort is: Remove unused dependencies, declare unlisted imports explicitly, and demote type-/test-only packages to devDependencies. The rest can wait behind it.
Evidence: priority ranking: top of 5 ranked by impact/effort
→ Remove unused dependencies, declare unlisted imports explicitly, and demote type-/test-only packages to devDependencies.
Architecture — module dependency matrix
1298 modules, 347 dependencies — 3 dependency cycles, shown as the red cell(s) above the diagonal. Rows and columns are the same modules, ordered so that a module only depends on ones above it. A cell means the row depends on the column, and its number is how many type pairs create that dependency. Read one thing: is anything above the diagonal? A mark there is a dependency cycle. (A cycle is all this shows — an unusual but cycle-free dependency sits below the diagonal like any other.)
Findings mapped to OWASP categories; the specific CVEs/secrets are in the Security dimension cards below and findings.md (redacted only on the public version of this report).
OWASP category
Findings
Severity
A03:2021 — Injection
50
High / Critical
A06:2021 — Vulnerable & Outdated Components
50
High / Critical
A02:2021 — Cryptographic Failures
42
High / Critical
A05:2021 — Security Misconfiguration
41
High / Critical
A04:2021 — Insecure Design
3
High / Critical
Roadmap
Begin by cleaning up the project's dependencies by removing unused packages and explicitly declaring all imports. Next, address the eight leaked secrets found in the codebase, focusing on run.sh, oauth.py, and public-build-values.json. Then, improve test coverage by adding tests for the 62.8% of production files that are currently unreachable. Finally, enhance data safety by enabling purge protection and soft-delete for critical resources, and ensure all interactive controls are fully keyboard-accessible.
Ranked by impact ÷ effort. "Helps" is the estimated gain on the 0–100 health score.
Do this
Helps
Effort
Dimension
Resolve the 8 Leaked secret finding(s) in Secret Scanning — start with run.sh (2), oauth.py (2), public-build-values.json.
Add a text alternative — alt on images (alt="" for purely decorative ones), a title or aria-label on meaningful svg, an aria-label or inner fallback content on canvas, and a captions <track> on video.
Keep a visible focus style (don't remove the outline without a replacement), guard animation with prefers-reduced-motion, and raise low-contrast colour pairs to at least 4.5:1.
OSV Dependency Vulnerabilities: High CVE: [GHSA redacted]
plugins/omi-ms365-app/requirements.txt
5.1
Mixed
OSV Dependency Vulnerabilities: High CVE: [GHSA redacted]
plugins/composio/requirements.txt
5.1
Mixed
OSV Dependency Vulnerabilities: High CVE: [GHSA redacted]
Methodology & how to trust this report
Watchdog is a deep, periodic assessment — run each sprint, monthly, or quarterly, taking the time to go wider and deeper than a quick check and surfacing in one coherent report what you'd otherwise piece together from a dozen separate tools. It scores deterministically: the same commit yields the same score, every run. 55 of 58 evaluated dimensions are computed purely by tools and static analysis (confidence 1.0); 3 documentation/naming judgement(s) are LLM-assisted and labelled advisory. Overall confidence is 0.9 — the weighted average across measured dimensions; it falls as more of the score leans on LLM-assisted judgement and rises when it's fully tool-backed.
Every figure here is one of three kinds, and we label which: ✓ Measured — a deterministic fact (LoC, complexity, coverage); ~ Modeled — an estimate from a stated model (cost, effort, value-at-risk), always a range with its assumptions, never a precise fact; ◐ Advisory — an LLM prose judgement. We never present a modelled estimate as if it were measured. Perfect or absent scores carry their provenance too (ADR-0011): ✓ Tool-verified means the property itself was measured across the surface; ○ Nothing flagged means the probes came back clean — a claim bounded by what a repository can show; ⊘ Not evidenced means a working control (a tested restore, an automated rollback) showed no positive evidence — absence of evidence is not evidence of a control, so it's excluded from the score rather than awarded a spurious 10; ◐ Sampled · advisory marks an LLM verdict over a bounded sample — advisory, never a deterministic measurement.
What we checked — 58 dimensions across the health lenses
Each chip is a dimension scored from real signals across architecture, testing, dependencies, security & compliance, documentation, git-history and code quality — in one coherent pass. A surface report typically covers a handful.
How to trust any code-health report — three questions
Can you open the finding? Real findings cite a repo-relative file and line you can open at the cited line — never an absolute scratch path. Here, 587 of 613 do; the remainder are repo-wide signals — a dimension-level measurement, not a single line. (Every path in this report is repo-relative by construction: paths are normalized at the producer and the report is rejected if any rooted path leaks through.)
Is there a tool behind the number? Every score below names the method that produced it — Roslyn, git, a scanner, or (for a handful of documentation/naming dimensions) an LLM labelled sampled · advisory — not a narrative.
Does re-running give the same result? Run it again on the same commit and the score — and this report, byte for byte — is identical. A report whose numbers move between runs is describing the run, not the code.
This report answers yes to all three. That's the bar to hold any assessment to.
Tools & methods
The actual versions used this run (captured at analysis time) — re-run on the same commit for the identical score.
Method
Backs
Version
Evaluator
Roslyn static analysis
Complexity, cohesion, coupling, dead code, API surface, layering
What ran differently this time — a tool absent, degraded, or that fell back to an estimate. Named openly, not folded silently into the scores. A degraded run also records its exact cause in diagnostics.md.
D30 Dependency Vulnerabilities — scanner not present in this environment — The backing tool was not installed where this scan ran, so this dimension was not scored. Install the tool (or run in the hosted environment, where it is always present) for a graded result.
Repo exclusion declarations: 50 pattern(s) declared (.gitattributes linguist-generated/vendored, .editorconfig generated_code) excluded 0 source file(s) from code-quality scoring. Declarations are the repo's own visible statement that a tree is machine-written or vendored — auditable in any diff, honored by GitHub the same way.
Limitations & what we did not check
Watchdog assesses the repository exactly as committed, and only the repository. By design it does not reach outside the source tree: the live cloud account, the running CI/CD pipeline, the host's branch-protection and approval rules, the production configuration, or a restore actually exercised against a backup are all out of scope. That boundary is a feature, not a gap — a repo-relative, deterministic scan re-runs identically on any commit and every finding opens at a real file and line, where a live audit can neither be reproduced nor traced. The visible consequence is that controls which leave no in-repo evidence are reported as "not evidenced" and excluded from the score rather than awarded a number a static scan cannot justify.
Per-dimension blind spots
For each dimension that was measured, what a static, repo-only scan structurally cannot see — the honest edge of the measurement, not a failure of it.
D4 Code Duplication: Duplication is token-similarity (jscpd) — it finds copy-paste, not semantic duplication expressed differently. Committed machine-written code (scaffolded migrations, designer/codegen output, protobuf/OpenAPI stubs, model snapshots) is EXCLUDED — its repetition is the tool's, not the team's — so the score reflects hand-written duplication only.
D5 Coupling: Coupling is measured between projects/assemblies — runtime coupling through DI, reflection, messaging or shared databases is invisible to a static reference graph.
D7 Architectural Integrity: Layering is checked against detected/declared rules — an architecture whose boundaries live in convention or in code review, not in a rule a scanner can read, is not enforced here.
D13 Secret Scanning: Secret detection is signature- and entropy-based on the current tree — a secret that does not match a known pattern, or one already rotated, will not be flagged (a clean scan is "nothing matched", not "no secrets exist").
D14 License Compliance: License compatibility is checked against declared package metadata and a policy — mislabelled or missing license metadata, and obligations that depend on how you distribute, are not resolved here.
D15 Churn × Complexity Hotspots: Churn hotspots come from git history — a freshly imported or squashed repository has no churn signal, and recent rewrites can mask a historically risky file.
D16 Bus Factor: Bus-factor is a time-decayed model of commit attribution (who has recently, repeatedly worked a file), not comprehension — pairing, review and reading-without-committing spread knowledge it can't see; bot commits and shared accounts still distort it.
D19 Documentation Quality: Documentation quality is judged by an LLM over a bounded sample of docs — it reads what is written, not whether the docs match the running system, and it is advisory, not a measurement.
D20 ADR Quality: ADR quality is an LLM read of the decision records present — it cannot know about decisions made and never recorded, and its verdict is sampled and advisory.
D21 Naming Consistency: Naming quality is an LLM judgement over a bounded sample — it assesses clarity/consistency of the names it sees, not domain-correctness, and is advisory.
D28 Secrets (history): Secret-history scanning sweeps the git log for known patterns — a secret that predates the available history, or never matched a signature, is not found (clean means "nothing matched in the history we can see").
D29 Static Analysis (SAST): SAST findings are pattern-based (semgrep) — it finds classes of bug it has rules for; logic flaws, auth/authorization gaps and issues needing runtime context are out of reach (and clean means "no rule matched").
D31 IaC & Container Security: IaC scanning checks Dockerfiles/Terraform/Kubernetes against best-practice rules — it cannot see the live cloud account, runtime configuration, or drift between the committed config and what is actually deployed.
D32 Data Compliance (PII/GDPR): PII/GDPR signals are heuristic pattern matches in code — they flag likely handling concerns, not legal compliance, and cannot trace where data actually flows at runtime.
D33 JS/npm Dependency Vulnerabilities: JS/npm CVE matching reads package manifests and lockfiles — risk from how a dependency is used, and advisories not yet published, fall outside this scan.
D34 Knowledge Freshness: Freshness is decayed commit RECENCY, not comprehension — code read often but rarely committed reads as orphaned, and stable code that genuinely needs no changes is penalised the same as forgotten code; bot/squash commits distort it like the bus factor.
D35 Change Coupling: Change coupling is co-change in COMMITS — files split across separate commits, or coupled only through a shared config/build step, read as uncoupled, and a sweeping commit (rename/format) is excluded so it doesn't couple everything. It shows that files change together, not WHY: a high coupling can be a healthy cohesive pair as readily as a hidden leak.
D40 Network Egress Confinement: Egress confinement is read from committed Kubernetes manifests — a policy applied out-of-band (cluster-default deny, a service mesh, or a cloud firewall/security group off-repo) is invisible, and a present NetworkPolicy is declared config, not proof the cluster admission-controller actually enforces it at runtime.
D41 Kernel & Syscall Confinement: Syscall/MAC confinement is read from committed manifests — a profile applied by a cluster-wide PodSecurity default or a mutating webhook off-repo isn't seen, and a declared seccomp/AppArmor profile is config presence, not proof the node's kernel actually loaded and enforced it.
D42 Runtime Threat Enforcement: Runtime enforcement is read from committed policy files — a Tetragon/Falco/Kyverno stack installed cluster-wide (Helm release, platform add-on) with no in-repo trace can't be credited, and a committed policy is declared intent, not proof the engine is running and blocking in the live cluster.
AC1 Text alternatives: Alt-text is detected structurally — the scan sees that an alternative EXISTS, not whether it meaningfully describes the image, and decorative-vs-missing is judged by attribute shape; runtime-injected images and a non-role=img decorative svg are out of scope. This is accessibility readiness, never a WCAG conformance claim.
AC2 Forms & labels: Label association is read from static markup — a label wired up at runtime (JS-set aria-labelledby, framework-injected ids) reads as missing, a present label says nothing about whether its text is correct. A known UI-library field component (e.g. a JSX <TextField>) is now checked conservatively — flagged only when it carries NO label/aria-label/aria-labelledby/id/name — but wrapper/context-labelled libraries (Chakra/Radix FormControl+FormLabel) aren't statically visible (possible false positive) and non-JSX lowercased components are still skipped. A clean result is "no unlabelled native control found", not a labelling proof.
AC3 Page structure: Page structure is read from the static markup tree — landmarks, headings and lang injected at runtime aren't seen, heading ORDER is checked structurally (not against the rendered visual hierarchy), and lang/title/main fire only on full documents, never partials, and the data-table check sees header-cell presence (a <th> exists), not whether each header correctly associates with its cells. Static readiness, not conformance.
AC4 Keyboard semantics: Keyboard semantics are inferred from markup attributes — interactivity wired purely in script, focus managed at runtime, and component-level handlers are invisible. A clean result means "no static keyboard-trap shape", not a keyboard-operability proof.
AC5 ARIA correctness: ARIA correctness is checked against the static role/attribute shape — roles/attributes set dynamically aren't seen, a valid role says nothing about whether it matches the element's real behaviour, and required-state checks are suppressed when a JSX spread could supply them.
AC6 Visual & motion safety: Contrast and motion safety are PARTIAL by construction — literal colours (hex/rgb/hsl/named) in inline styles, in-repo <style> blocks, in-repo .css files, var() tokens, Tailwind neutral utilities and CSS-in-JS top-level declarations are read (same-rule/same-element colour+background pairs only); computed/runtime/theme colour, external-CDN stylesheets, CSS-in-JS dynamic (${…}) and nested-selector colours, cross-element pairs and image contrast stay out of reach, so a clean result is bounded by what the static CSS itself shows.
AC7 A11y enforcement: Enforcement is scored from in-repo config/CI evidence only — an a11y gate enforced in external tooling with no in-repo trace can't be credited, and a configured linter is presence, not proof the rules actually run or block a merge.
AX9 CQS / query purity: Handlers are found by interface/name convention — a query handler using neither is not seen. Mutation is a resolved write/publish invocation (SaveChanges/repository/bus), so a write hidden behind a hand-rolled wrapper, reflection, or a string-keyed service locator resolves to a non-persistence type and isn't flagged; it detects that a query writes state, not whether the write is a legitimate read-side cache update. Clean means "no resolved write/publish in a query body", not a proof of CQS purity.
DM4 Rich vs anemic model: Behaviour is detected as state mutation inside a method body — a method that enforces an invariant by validating-and-throwing without mutating reads as a query, and mutation delegated through an interface the scan can't resolve isn't credited; entities with zero public properties still drop out of the population. It detects that state changes, not whether the rule is correct.
DM6 Domain ↔ infrastructure boundary: Infrastructure reached through a hand-rolled wrapper, a domain-named facade, reflection, or a string-keyed service locator resolves to a non-infra type and isn't seen; the body scan is symbol resolution over syntax, not full dataflow. A clean result means "no resolved infra reference in a domain body", not a proof of purity.
M4 Documentation accuracy: Onboarding quality is an LLM read of the docs/setup present — it cannot run the onboarding or measure how long a real new joiner takes; the verdict is sampled and advisory.
P4 Deployment & Rollback: Approval/branch-protection rules live in repository settings the scan cannot see — only their in-repo evidence (config files, workflows) is checked, so a control enforced purely in the host's settings reads as "not evidenced".
P5 DR & Backup: Backup/restore and disaster-recovery readiness is judged from in-repo evidence — a config that exists is not a tested restore, so the absence of positive evidence is reported as "not evidenced", never scored as present.
P6 Release Hygiene: Rollback/observability controls are inferred from repo artefacts (pipelines, dashboards-as-code) — controls configured in external tooling, with no in-repo trace, cannot be credited.
The LLM boundary
LLM-set scores this run (4): D19, D20, D21, M4 (model: Local LLM). For these, a model reads a bounded sample and sets the numeric score (documentation, ADR quality, naming, comment value, onboarding) — D25 sets the ADR-conformance fraction over sampled code, D22 judges API accuracy over a sample. These are sampled and advisory by design: they vary at the margins between runs and are never a deterministic measurement. Every other score in this report is tool-computed at confidence 1.0.
What it measures: Copy-pasted code that should be shared instead.
Method: Code duplication via token-stream sliding windows with type-aware normalization (locals masked, type names preserved), density-scored per KLoC of production code. Deterministic.
+ 41 more group(s) — more in Appendix A; the complete list is findings.md.
✓ On the Gold path — maintain.
Detailed fixes: d4_recommendation.md · top locations in Appendix A, every location in findings.md.
Do you agree with this assessment?
D5 · Coupling6.9 / 10Adequate✓ Tool-verified
What it measures: Whether volatile projects sit underneath others that depend on them (so their churn ripples upward), and whether project dependencies form cycles. A widely-depended-on but stable shared/kernel project is healthy, not penalised.
Method: Dependency cycles via elementary-DFS over real .csproj references, plus Martin instability (afferent/efferent) per project. Exhaustive over the reference graph, deterministic.
Coverage: Exhaustive · type-level: afferent/efferent coupling + cycles computed over every production type — the population is all types, not a name convention.
Enforce Coupling in CI to reach Verified (currently Documented). — Hardens enforcement from Documented toward Prevented — provenance only; does not change the score.
What it measures: Whether the code respects its intended layering / architecture rules.
Method: Enforcement rung (Prevented/Verified/Documented) per checkable ADR via Roslyn, plus dependency cycles via the engine shared with D5/AX3. Deterministic, exact.
What it measures: Whether any secrets (keys, tokens, passwords) have leaked into the code.
Method: In-process native secret scanner (entropy plus signature patterns) across all tracked files; no external tool. A clean result is a measured 10, not no-data zero. Deterministic.
Resolve the 8 Leaked secret finding(s) in Secret Scanning — start with run.sh (2), oauth.py (2), public-build-values.json. — One of this dimension's main actionable groups (8 issue-level).
Enforce Secret Scanning in CI to reach Verified (currently Documented). — Hardens enforcement from Documented toward Prevented — provenance only; does not change the score.
Detailed fixes: d13_recommendation.md · top locations in Appendix A, every location in findings.md.
What it measures: Whether the licenses of third-party packages are compatible with your policy.
Method: Third-party package licenses resolved from declared package metadata and checked against the configured policy (allow/deny/copyleft). Deterministic; clean = no incompatible license found at metadata depth.
What it measures: Files that change often and are also complex — the riskiest hotspots.
Method: Per production file churn times cyclomatic complexity over a rolling window, computed from git and Roslyn/JS/Razor analysis. Exhaustive, deterministic per commit date.
Detailed fixes: d15_recommendation.md · top locations in Appendix A, every location in findings.md.
Do you agree with this assessment?
D16 · Bus Factor8.1 / 10Strong✓ Tool-verified
What it measures: Whether knowledge is concentrated in too few people (the "bus factor").
Method: Living knowledge per author via time-decayed commit attribution (6-month half-life, focus weighting) across largest source files. Deterministic, avoids blame's mechanical-refactor false positives.
489 source file(s) have their living knowledge concentrated in one author (≥90% of recent, decayed contribution). The largest is desktop/macos/agent/src/index.ts.
Off-boarding risk: anonymized user #1 · ×14
Further sole-owners (lower concentration)
What to do
Resolve the 14 Off-boarding risk finding(s) in Bus Factor. — One of this dimension's main actionable groups (14 recommendation-level).
Resolve the 1 Further sole-owners (lower concentration) finding(s) in Bus Factor. — One of this dimension's main actionable groups (1 recommendation-level).
Detailed fixes: d16_recommendation.md · top locations in Appendix A, every location in findings.md.
What it measures: Whether the project's documentation is clear, complete, and useful.
Method: Judged by language model at low temperature (0.0-0.1) on a deterministic doc sample (READMEs plus first 25 architecture docs), with two-pass stability filtering. Advisory, sampled.
Omi's documentation is comprehensive, well-organized across READMEs, architecture/design docs, and an extensive XML-doc coverage. It starts with a strong branding banner showing the site-wide links (Discord, GitHub stars, license), then dives into Quick Start for macOS and Windows, prerequisites, and a full local setup guide that clips mid-documentation before covering API endpoints, architecture, hardware, and more. The plugins example project is also well written, complete with prerequisites, installation, running steps, and a detailed feature list. Despite the clipped sections, every named heading in the outline exists (omi; Quick Start; macOS; Windows; Mobile App; Documentation; Getting Started; Building Apps; API & SDKs; Architecture; Omi Hardware; License) and is not flagged as missing. The Omi SDK documentation is comprehensive and well-structured across READMEs for Swift, React Native, Python, a CLI tool, device BLE protocol packages, and two migration tools (Mixpanel to archive, Uber call app). Each file is an auto-generated markdown doc that starts with an overview card group showing the language/platform it supports, then dives into installation, quick start, configuration, and related topics. The docs are clipped mid-quick-start section in most files but the outline lists all named sections (Installation; Quick Start; Auth; Profiles; Local Omi Desktop API; Command surface; Global flags; Exit codes; For agents; Rate limits; Development; Editable install with dev extras; Run the test suite; Lint; Build a wheel + sdist) and are not flagged as missing. The content is clear, complete for its audience, and well-organized. This project is a collection of Omi integration apps for USGS, Twitter/X, Stack Overflow, Slack, and Shopify. Each README is well written: it states what the app does, lists configuration fields with defaults, shows example prompts, provides local development commands, and documents environment variables. The architecture/design docs (OMI Chat Tools manifest, Railway deployment) are also present but not shown in this summary. All visible sections appear in the outline — no missing section to flag.
The BLE setup warning (pod install after npm install) and platform-specific XML config are clipped mid-quick-start.sdks/react-native/README.md
The Omi App Configuration table lists Setup URL and Setup Completed URL, but the README does not show how to navigate from the setup page onward (e.g., after auth, where to click to complete setup).plugins/omi-twitter-chat-tools-app/README.md
The README describes voice-activated Twitter posting but does not mention how to handle rate limits and token refresh after one-time auth.plugins/omi-twitter-app/README.md
✓ On the Gold path — maintain.
Detailed fixes: d19_recommendation.md · top locations in Appendix A, every location in findings.md.
What it measures: Whether architecture decisions are recorded well (context, decision, consequences).
Method: Per-ADR judgment by language model at low temperature with two-pass stability; confidence is share of ADRs evaluated; enforcement-field presence detected deterministically. Advisory.
What it measures: Whether names — types, methods, variables — are clear and consistent.
Method: Judged by language model at low temperature (0.0-0.1) on a deterministic random symbol sample (fixed size, not exhaustive), with disclosed confidence band. Advisory, sampled.
What it measures: Whether any secrets were ever committed — scanned across the full git history, not just now.
Method: Git-history secret scan via gitleaks detect over full history in an isolated checkout; each match flagged High. Exhaustive; degrades cleanly when tool absent.
40 finding(s): 0 critical, 40 high, 0 medium, 0 low. Remediation for historically-committed secrets is credential rotation — they remain in history regardless of later deletion.
Secret: generic-api-key · ×33backend/runtime_images.json:136detected by gitleaks finding
Rotate the exposed credentials — git history can't be un-committed
What to do
Resolve the 33 Secret finding(s) in Secrets (history) — start with public-build-values.json (3), firebase_options.dart (3), run.sh (2). — One of this dimension's main actionable groups (33 issue-level).
Resolve the 1 Rotate the exposed credentials finding(s) in Secrets (history). — One of this dimension's main actionable groups (1 recommendation-level).
Detailed fixes: d28_recommendation.md · top locations in Appendix A, every location in findings.md.
What it measures: Real static-analysis (SAST) findings — likely security bugs in the code, any language.
Method: Polyglot static analysis via semgrep across the repo using the pinned, image-baked p/security-audit + p/owasp-top-ten rulesets (no scan-time registry fetch); severity rules (ERROR/WARNING/INFO) map to a full-band severity-weighted score. Exhaustive, deterministic; degrades on parse failure.
Coverage: semgrep pattern rules over all files — exhaustive for the rule set, blind to classes of bug without a rule (clean = no rule matched).
High: run-shell-injection · ×50.github/actions/deploy-backend-stack/action.yml:78detected by semgrep finding
What to do
Resolve the 50 High finding(s) in Static Analysis (SAST) — start with action.yml (50). — One of this dimension's main actionable groups (50 issue-level).
Detailed fixes: d29_recommendation.md · top locations in Appendix A, every location in findings.md.
High IaC: DS-0029 · ×30backend/Dockerfiledetected by trivy finding
Medium IaC: DS-0013 · ×11backend/Dockerfiledetected by trivy finding
What to do
Resolve the 30 High IaC finding(s) in IaC & Container Security — start with Dockerfile (19), Dockerfile.notifications_job (2), Dockerfile.datadog (2). — One of this dimension's main actionable groups (30 issue-level).
Resolve the 11 Medium IaC finding(s) in IaC & Container Security — start with prod_agent_vm_reaper_cronjob.yaml (4), dev_omi_backend_listen_cronjob.yaml (4), prod_omi_backend_listen_cronjob.yaml (2). — One of this dimension's main actionable groups (11 warning-level).
Detailed fixes: d31_recommendation.md · top locations in Appendix A, every location in findings.md.
Do you agree with this assessment?
D32 · Data Compliance (PII/GDPR)5.7 / 10Adequate✓ Tool-verified
What it measures: Likely personal-data (PII / GDPR) handling concerns — logging or storing data without safeguards.
Method: Heuristic PII/GDPR pattern scan via semgrep across the repo, using Watchdog's own ruleset (personal data reaching log/console sinks, URLs and query strings, or unprotected browser storage); matches map to severity and a 0-10 wide normalizer. NotApplicable when the scan runs and detects no PII/GDPR surface (no unearned 10); reported LOUDLY as a measurement gap, never as not-applicable, if the ruleset is missing from the analyzer image. Exhaustive, advisory-leaning; degrades on parse failure.
High: watchdog-sensitive-personal-data-in-logdesktop/windows/src/main/glow/glowWindow.ts:186detected by semgrep finding
Medium: watchdog-personal-data-in-log · ×2backend/routers/mcp.py:242detected by semgrep finding
What to do
Resolve the 2 Medium finding(s) in Data Compliance (PII/GDPR) — start with mcp.py, twilio_service.py. — One of this dimension's main actionable groups (2 warning-level).
Resolve the 1 High finding(s) in Data Compliance (PII/GDPR) — start with glowWindow.ts. — One of this dimension's main actionable groups (1 issue-level).
Detailed fixes: d32_recommendation.md · top locations in Appendix A, every location in findings.md.
What it measures: Whether anyone still has living knowledge of each file, or it has been orphaned — last understood long ago by someone now gone quiet. The sibling of the bus factor: D16 asks who owns it, D34 asks whether anyone still knows it.
Method: File orphaning as total living-knowledge decay below one focused-commit's worth within a year, computed per-file from the D16 decay model. Exhaustive, deterministic over fixed history.
What it measures: Whether files that change together actually belong together — pairs that repeatedly co-change in git history despite having no explicit code dependency, surfacing the hidden/logical coupling (and boundaries in the wrong place) a static scan can't see.
Method: Pairwise co-occurrence over the per-commit file sets in git history (production source only — tests and generated dropped): Degree-of-Coupling = shared ÷ min individual revisions, reported above noise floors (each file ≥10 revisions, ≥5 shared commits, ≥50% strength); sweeping commits excluded. Deterministic over fixed history.
Coverage: Population: PRODUCTION source files only — test and generated files are dropped before pairing, so a class co-changing with its own test (trivially ~100%) can't drown the real production↔production coupling. Pairs ranked by Degree-of-Coupling; coupling through a build step, config, or non-source file isn't seen.
What it measures: Whether the build pipeline provides supply-chain integrity — generated provenance/attestation, signed artifacts (cosign/sigstore), an SBOM, and pinned build actions. Presence of the configuration, not a runtime guarantee.
Method: Supply-chain provenance/signing read deterministically from CI/build config (.github/workflows, .gitlab-ci.yml, azure-pipelines, Jenkinsfile, .circleci) + the release surface: four signals — generated provenance/attestation (SLSA/in-toto/actions-attest), artifact signing (cosign/sigstore/gitsign), an SBOM (syft/sbom-action/*.spdx.json/*.cdx.json), and SHA-pinned build actions — scored 10·present/denom. NotApplicable without a build pipeline. Detects configuration presence, not runtime enforcement.
Resolve the 1 Unpinned build actions finding(s) in Supply-chain Provenance & Signing. — One of this dimension's main actionable groups (1 warning-level).
Resolve the 1 PR-triggered workflow without a permissions block finding(s) in Supply-chain Provenance & Signing. — One of this dimension's main actionable groups (1 warning-level).
Resolve the 1 Secret passed as a command-line argument finding(s) in Supply-chain Provenance & Signing. — One of this dimension's main actionable groups (1 warning-level).
Detailed fixes: d36_recommendation.md · top locations in Appendix A, every location in findings.md.
What it measures: Whether the repository publishes a coordinated-vulnerability-disclosure policy (SECURITY.md or security.txt) with a reporting contact, so finders know how to report a vulnerability. Presence of a policy file with a contact, not whether the policy is adequate or honoured.
Method: Vulnerability-disclosure policy read deterministically from the repo: a SECURITY.md (root/.github/docs) or .well-known/security.txt / security.txt, regex-checked for a reporting contact (email / URL / mailto). Present + contact → 10; present without a contact → 4; NotApplicable when no policy file exists (it may live off-repo). Detects the policy file's presence + contact, not its adequacy.
What it measures: Whether dependencies have known published vulnerabilities (CVEs) per the OSV database — read natively from whatever lockfile the repository ships (Cargo, npm, Go, Python, Maven, RubyGems, …). D33 and D30 add ecosystem-specific scanners on top for npm and .NET.
Method: Multi-ecosystem dependency-CVE scan via osv-scanner --recursive (queries the osv.dev database + parses lockfiles natively across ecosystems: npm package-lock/yarn/pnpm/bun, Go go.mod, Rust Cargo.lock, Maven/Gradle pom.xml/gradle.lockfile, PyPI requirements.txt/poetry.lock/Pipfile.lock, Composer composer.lock, RubyGems Gemfile.lock, Hex mix.lock, pub pubspec.lock, Swift Package.resolved); severity tally (Critical/High/Medium/Low) to 0-10 tight normalizer (8.0). NotApplicable only when the repo declares no supported non-.NET dependency lockfile (a NuGet-only repo stays NotApplicable — .NET CVEs are D30's domain); coverage needs a resolved lockfile. Additive to D33 (trivy fs); exhaustive + deterministic, DB kept fresh.
High CVE: [GHSA redacted] · ×31backend/diarizer/requirements.txtdetected by osv-scanner finding
High vulnerability: [GHSA redacted] · ×4backend/agent-proxy/requirements.txtdetected by osv-scanner finding
Critical CVE: [GHSA redacted]plugins/requirements.txtdetected by osv-scanner finding
Medium CVE: [GHSA redacted] · ×13sdks/react-native/package-lock.jsondetected by osv-scanner finding
Medium vulnerability: [GHSA redacted]plugins/requirements.txtdetected by osv-scanner finding
What to do
Resolve the 31 High CVE finding(s) in OSV Dependency Vulnerabilities — start with requirements.txt (15), package-lock.json (10), pnpm-lock.yaml (6). — One of this dimension's main actionable groups (31 issue-level).
Resolve the 4 High vulnerability finding(s) in OSV Dependency Vulnerabilities — start with requirements.txt (3), pnpm-lock.yaml. — One of this dimension's main actionable groups (4 issue-level).
Resolve the 1 Critical CVE finding(s) in OSV Dependency Vulnerabilities — start with requirements.txt. — One of this dimension's main actionable groups (1 issue-level).
Detailed fixes: d38_recommendation.md · top locations in Appendix A, every location in findings.md.
What it measures: Whether Kubernetes workloads restrict network EGRESS with a NetworkPolicy (or Cilium policy), limiting where a compromised pod can send data or reach a command-and-control server. Presence of committed egress-restricting policy, not runtime enforcement.
Method: Deterministic YAML-manifest inspection (no external tool, no Roslyn — language-agnostic): Kubernetes workloads gate applicability; credits a NetworkPolicy / Cilium policy that restricts egress (policyTypes: [Egress] / egress rules). Reward-leaning (neutral floor climbing to 10, never a deduction — baseline misconfigs stay with D31). Deterministic.
What it measures: Whether Kubernetes workloads confine the kernel boundary — a seccomp profile (RuntimeDefault/Localhost) plus an AppArmor/SELinux mandatory-access-control layer — shrinking the syscall attack surface a container escape would use. Presence of committed confinement config, not runtime enforcement.
Method: Deterministic YAML-manifest inspection (no external tool, no Roslyn): on Kubernetes workloads, credits a seccomp profile (RuntimeDefault/Localhost) and an AppArmor/SELinux MAC layer. Reward-leaning (neutral floor climbing to 10); NotApplicable without workloads. Deterministic.
What it measures: Whether the Kubernetes deployment wires runtime threat enforcement — a detection engine (Tetragon/Falco) and/or an admission-control policy gate (Kyverno / OPA Gatekeeper / PodSecurity). Presence of committed policy, not a runtime guarantee.
Method: Deterministic YAML-manifest inspection (no external tool, no Roslyn): on Kubernetes workloads, credits a runtime threat-detection engine (Tetragon TracingPolicy / Falco) and an admission-control policy (Kyverno / OPA Gatekeeper / PodSecurity). Reward-leaning; NotApplicable without workloads. Deterministic.
Resolve the 1 No runtime threat detection finding(s) in Runtime Threat Enforcement. — One of this dimension's main actionable groups (1 recommendation-level).
Resolve the 1 No admission-control policy finding(s) in Runtime Threat Enforcement. — One of this dimension's main actionable groups (1 recommendation-level).
Detailed fixes: d42_recommendation.md · top locations in Appendix A, every location in findings.md.
Do you agree with this assessment?
Frontend & cross-cutting dimensions
R = React/JS · M = Maturity · P = Readiness.
AC1 · Text alternatives5.2 / 10Adequate✓ Tool-verified
Other · Accessibility — Whether non-text content carries a text alternative — img/area/input[type=image] have alt, a meaningful svg has a title or aria-label, video has a captions track, and object/embed/canvas have a name or fallback content. Static markup readiness, not a WCAG conformance claim.
Method: Static markup-model scan: every img/area/input[type=image] checked for alt, svg[role=img] for a title/aria-label, video for a captions <track>. Components skipped, spreads suppressed. Deterministic, hard fact per element.
A <canvas> with no aria-label/title and no inner fallback content can't be described by assistive tech. Add a name or fallback content. (×2) — index.html:8, BrainMap.tsx:119
An image with no alt (and no aria-label/aria-labelledby) is unreadable to assistive tech. Add alt — alt="" if it's purely decorative. (×6) — okpage.html:89, setup_notion_crm.html:96, setup_notion_crm.html:97, …
What to do
Add a text alternative — alt on images (alt="" for purely decorative ones), a title or aria-label on meaningful svg, an aria-label or inner fallback content on canvas, and a captions <track> on video.
Do you agree with this assessment?
AC2 · Forms & labels5.3 / 10Adequate✓ Tool-verified
Other · Accessibility — Whether form controls have a programmatic label (an associated label, aria-label or aria-labelledby), buttons have text, links have an accessible name, fieldsets have a non-empty legend, known UI-library field components carry a label prop, and a placeholder isn't used as the only label. Static markup readiness, not a WCAG conformance claim.
Method: Static markup-model scan: inputs/selects/textareas checked for an associated label[for]/wrapping label/aria-label/aria-labelledby (per document), buttons for accessible text, fieldsets for a legend; placeholder-only labelling flagged. Deterministic, hard fact per control.
This control has only a placeholder — a placeholder is not a label (it vanishes on input and many AT ignore it). Add a <label htmlFor>, a wrapping <label>, or aria-label. (×19) — AddReviewDialog.tsx:117, BarChatSurface.tsx:199, ChatHistoryPopover.tsx:73, …
This control has no associated label. Add a <label htmlFor> / wrapping <label> / aria-label / aria-labelledby so assistive tech can name it. (×5) — HistorySessionRow.tsx:93, ConversationListRow.tsx:156, PageHeader.tsx:63, …
This UI-library field component has no label / aria-label / aria-labelledby / id / name — and neither does anything it renders — so it likely renders an unlabelled control. Name it whichever way this library supports: a label prop, an aria-label, or an id on the rendered control with a <label htmlFor> pointing at it. For a group of controls, name the group itself (aria-label, or a fieldset with a legend) — labelling each item leaves the set unnamed. — FontSizeCard.tsx:50
What to do
Give every control a programmatic label (a <label for> / wrapping <label> / aria-label) and every button text — a placeholder is not a label.
Other · Accessibility — Whether pages declare a language (well-formed BCP-47) and a non-empty title, expose exactly one main landmark and a sane heading order with non-empty headings, keep zoom enabled, title their iframes, give data tables header cells, and avoid meta-refresh. Static markup readiness, not a WCAG conformance claim.
Method: Static markup-model scan: html lang, document <title>, a main landmark and heading order on full documents only, plus zoom-disabling viewports, untitled iframes and meta-refresh anywhere. Deterministic, per structural checkpoint.
user-scalable=no/0 or a maximum-scale below 2 stops low-vision users zooming to 200%. Remove the zoom restriction from the viewport meta. — index.html:55
No <main> (or role="main") means no "skip to content" target and a weaker landmark map. Wrap the primary content in <main>. (×9) — auth_callback.html:2, index.html:2, oauth_callback.html:2, …
The page declares no language, so assistive tech can't pick the right pronunciation. Add lang (e.g. lang="en"). (×9) — index.html:2, oauth_callback.html:2, index.html:2, …
A timed meta-refresh moves users with no control or warning. Use a server redirect or a user-triggered action instead. — oauth_callback.html:6
An empty heading appears in the document outline announced with no text. Give the heading text, or remove it if it's decorative. — base.html:208
Skipping heading levels breaks the document outline assistive tech relies on. Don't jump levels — increase by at most one. (×4) — dashboard.html:211, setup.html:458, setup.html:476, …
What to do
Declare <html lang>, a document <title> and a <main> landmark, keep headings in order, leave zoom enabled, title iframes and drop meta-refresh.
Other · Accessibility — Whether interactive behaviour is keyboard-reachable — no click handler on a non-interactive element lacking a role, tabindex and key handler, no positive tabindex, no href-less anchor, no placeholder-href (#/javascript) link acting as a button. Static markup readiness, not a WCAG conformance claim.
Method: Static markup-model scan: click handlers on non-interactive elements lacking role+tabindex+key handler, positive tabindex values, and href-less anchors. Components skipped, spreads suppressed. Deterministic, hard fact per element.
A click handler on a plain element with no role and no tabindex: even where another element's key handler can reach it, assistive tech can neither focus it nor announce what it is. Use a <button>, or add role + tabIndex={0}. (×2) — HistorySessionRow.tsx:67, SettingsPage.tsx:327
A click handler on a plain element isn't keyboard-operable. Use a <button>, or add role + tabIndex={0} + a key handler. (×19) — ConversationRowContextMenu.tsx:124, DateFilterButton.tsx:66, MoveToFolderMenu.tsx:98, …
A click handler on a plain element isn't keyboard-operable. Use a <button>, or add role + tabindex="0" + a key handler. — setup_zapier.html:118
What to do
Make custom controls keyboard-operable (role + tabindex + key handler), drop positive tabindex, and give anchors a real href.
Other · Accessibility — Whether ARIA is used correctly — valid non-abstract roles, the ARIA state a role requires, valid (non-misspelled) aria-* attribute names, in-enum values for token-typed aria-* attributes, and no aria-hidden on (or wrapping) a focusable element. Static markup readiness, not a WCAG conformance claim.
Method: Static markup-model scan: role values checked against the WAI-ARIA role set (abstract/invalid flagged), required ARIA state for a role, and aria-hidden on a focusable element. Deterministic, role/attribute level.
Other · Accessibility — Whether focus outlines aren't removed without a replacement, motion respects prefers-reduced-motion, and literal CSS colour pairs meet contrast — PARTIAL: inline styles, in-repo <style> blocks, in-repo .css files, var() tokens, Tailwind neutral utilities and CSS-in-JS literals are read (hex/rgb/hsl/named), never computed/runtime/external-CDN colour. Static markup readiness, not a WCAG conformance claim.
Method: Static markup/CSS scan: inline outline:none/0, literal inline colour/background contrast against the 4.5:1 AA floor, and <style>-block animation without a prefers-reduced-motion guard. Deterministic but PARTIAL — only inline styles and in-repo CSS literals are visible.
`.status-badge.completed` sets color: white on background-color: #10B981 — 2.5:1, below the 4.5:1 WCAG AA minimum for normal text. Darken or lighten one of them. — base.html:7
`.home-button` sets color: white on background-color: #007bff — 4.0:1, below the 4.5:1 WCAG AA minimum for normal text. Darken or lighten one of them. — notion_success.html:5
`button` sets color: white on background-color: #4CAF50 — 2.8:1, below the 4.5:1 WCAG AA minimum for normal text. Darken or lighten one of them. — index.html:7
This element's Tailwind text/background colour utilities fall below the 4.5:1 WCAG AA minimum for normal text. Pick a darker or lighter shade. (×22) — page.tsx:1446, loading.tsx:6, loading.tsx:14, …
What to do
Keep a visible focus style (don't remove the outline without a replacement), guard animation with prefers-reduced-motion, and raise low-contrast colour pairs to at least 4.5:1.
Other · Accessibility — Whether accessibility is ENFORCED in the toolchain — an accessibility checker configured over the markup (an a11y lint rule set, e.g. eslint-plugin-jsx-a11y or vuejs-accessibility where the project lints JavaScript) and an automated accessibility assertion wired into tests or CI (axe/pa11y/Lighthouse or an equivalent) — on the Documented→Verified→Prevented ladder.
Method: Repo config/CI scan: an accessibility checker configured over the markup (an a11y lint rule set such as eslint-plugin-jsx-a11y / vuejs-accessibility where JavaScript is linted) and an automated accessibility assertion in tests or CI (axe/pa11y/Lighthouse or equivalent), graded on the Documented→Verified→Prevented rungs. Deterministic, presence/rung detection.
An a11y linter is configured (Documented) but there's no automated a11y test or CI gate. Add vitest-axe to verify at runtime, then gate it in CI.
What to do
Accessibility linting is already configured — assert with vitest-axe in tests, then gate axe/pa11y/Lighthouse in CI.
Other · Architecture — Whether the project-reference graph is acyclic (cycles block independent build/deploy and signal eroding boundaries).
Method: Project reference cycles via elementary-DFS over real .csproj references, using the engine shared with D5/D7; cyclic versus acyclic. Exhaustive, deterministic.
Other · Architecture — Whether dependencies point inward (Domain ← Application ← Infrastructure/Web) — the clean-architecture dependency rule, checked across the project graph.
Method: Layer violations by name-segment inference (Domain/Core to Application to Infrastructure/Web) over the project-reference graph. Exhaustive over all projects, deterministic.
Other · Architecture — Whether read (query) handlers stay side-effect-free — a query that writes persistent state or raises events breaks CQS and makes reads unsafe to retry, cache, or route to a read replica.
Method: Roslyn scan: CQRS handlers classified query-vs-command by interface (IQueryHandler/ICommandHandler/IRequestHandler<TQuery,TResult>) and name convention (*Query/Get*/Find* vs *Command); each query handler's body checked for persistent-state writes (SaveChanges/repository Add-Update) or event publishes by resolved invocation. Deterministic, type-level, exhaustive over the detected handlers.
Coverage: Population: CQRS handlers identified by IQueryHandler/ICommandHandler/IRequestHandler interface + *Query/Get*/Find*/*Command NAME convention; query purity then checked exhaustively within that set — a query handler using neither convention is invisible, and mutation is a resolved persistence/publish CALL, not full dataflow.
Other · Domain Modelling — Whether aggregates reference each other by identity (id) rather than by direct object reference — the core DDD consistency-boundary rule.
Method: Roslyn (DDD-gated): aggregate roots identified by convention; each aggregate field checked for direct references to other aggregates versus id-only. Deterministic, DDD-native.
Coverage: Population: aggregate roots identified by AggregateRoot/IAggregateRoot base/interface NAME convention; reference-by-identity then checked exhaustively within that set — a root not using those names is invisible.
`EmulatorView` references the aggregate root `PeripheralCatalog` directly (via `catalog`) — hold its `PeripheralCatalogId` instead. — main.rs:12
`EmulatorView` references the aggregate root `FlashSession` directly (via `flash`) — hold its `FlashSessionId` instead. — main.rs:19
What to do
Reference other aggregates by their strongly-typed id, never by object reference, so each aggregate stays an independent consistency boundary.
Do you agree with this assessment?
DM4 · Rich vs anemic model3.0 / 10Weak✓ Tool-verified
Other · Domain Modelling — Whether aggregates/entities carry the behaviour that protects their invariants, rather than being data bags driven by external services.
Method: Roslyn (DDD-gated): entity method BODIES classified mutator-vs-query — only methods that mutate the entity's own declared state count as invariant-protecting behaviour, so a getter/passthrough doesn't rescue an anemic class. Deterministic, exhaustive over domain-layer entities.
Coverage: Population: entities by name/base convention; rich-vs-anemic judged by classifying each method body mutator-vs-query — logic-bearing domain types outside the convention are invisible.
`ProactiveIntent` is an aggregate/entity with 1 data propert(ies) but no state-changing behaviour (only data and queries) — the business logic lives in a service. — chat_first.py:124
`WhisperTranscriber` is an aggregate/entity with 1 data propert(ies) but no state-changing behaviour (only data and queries) — the business logic lives in a service. — lib.rs:59
What to do
Move business rules onto the aggregates/entities they govern so invariants are enforced at the source, not in anemic services.
Other · Domain Modelling — Whether entities protect their state (private/init-only setters) instead of exposing public setters that bypass invariants. Softened when a rehydration framework (Marten/EF) is present.
Method: Roslyn (DDD-gated): public setters on entities detected; score softened when Marten/EF rehydration frameworks present. Deterministic, framework-aware.
Coverage: Population: entities by convention; encapsulation (setter shape) checked exhaustively within the set.
Other · Domain Modelling — Whether the domain layer stays free of infrastructure dependencies (EF/Marten/HTTP/ASP.NET) — the clean-architecture dependency rule.
Method: Roslyn (DDD-gated): domain-layer types scanned for infrastructure usage in member SIGNATURES and inside method/accessor BODIES — resolved calls and object-creations into EF/Marten/HTTP/Mongo/Redis/message-bus types (not just a namespace allowlist). Deterministic, symbol-resolved, exhaustive over domain-layer bodies, DDD-native.
Coverage: Domain layer identified by NAMESPACE heuristic; infrastructure then resolved by symbol in member SIGNATURES and method/accessor BODIES — rename the layer and the check evaporates.
`IndexedFileRecord` is a domain concept fused to persistence ON ITS OWN DECLARATION (the GRDB `PersistableRecord`/`FetchableRecord` conformance), with no separate pure-domain layer — the domain entity IS the ORM row (active-record). A schema/persistence change ripples straight into the domain type. Keep the ORM/table mapping on a boundary adapter and the domain entity framework-free (or map row↔domain through a repository). — IndexedFileRecord.swift:48
`LocalKGNodeRecord` is a domain concept fused to persistence ON ITS OWN DECLARATION (the GRDB `PersistableRecord`/`FetchableRecord` conformance), with no separate pure-domain layer — the domain entity IS the ORM row (active-record). A schema/persistence change ripples straight into the domain type. Keep the ORM/table mapping on a boundary adapter and the domain entity framework-free (or map row↔domain through a repository). — KnowledgeGraphRecord.swift:6
`LocalKGEdgeRecord` is a domain concept fused to persistence ON ITS OWN DECLARATION (the GRDB `PersistableRecord`/`FetchableRecord` conformance), with no separate pure-domain layer — the domain entity IS the ORM row (active-record). A schema/persistence change ripples straight into the domain type. Keep the ORM/table mapping on a boundary adapter and the domain entity framework-free (or map row↔domain through a repository). — KnowledgeGraphRecord.swift:46
`LiveNoteRecord` is a domain concept fused to persistence ON ITS OWN DECLARATION (the GRDB `PersistableRecord`/`FetchableRecord` conformance), with no separate pure-domain layer — the domain entity IS the ORM row (active-record). A schema/persistence change ripples straight into the domain type. Keep the ORM/table mapping on a boundary adapter and the domain entity framework-free (or map row↔domain through a repository). — LiveNoteModels.swift:23
`AIUserProfileRecord` is a domain concept fused to persistence ON ITS OWN DECLARATION (the GRDB `PersistableRecord`/`FetchableRecord` conformance), with no separate pure-domain layer — the domain entity IS the ORM row (active-record). A schema/persistence change ripples straight into the domain type. Keep the ORM/table mapping on a boundary adapter and the domain entity framework-free (or map row↔domain through a repository). — AIUserProfileService.swift:7
`ActionItemRecord` is a domain concept fused to persistence ON ITS OWN DECLARATION (the GRDB `PersistableRecord`/`FetchableRecord` conformance), with no separate pure-domain layer — the domain entity IS the ORM row (active-record). A schema/persistence change ripples straight into the domain type. Keep the ORM/table mapping on a boundary adapter and the domain entity framework-free (or map row↔domain through a repository). — ActionItemModels.swift:8
`StagedTaskRecord` is a domain concept fused to persistence ON ITS OWN DECLARATION (the GRDB `PersistableRecord`/`FetchableRecord` conformance), with no separate pure-domain layer — the domain entity IS the ORM row (active-record). A schema/persistence change ripples straight into the domain type. Keep the ORM/table mapping on a boundary adapter and the domain entity framework-free (or map row↔domain through a repository). — ActionItemModels.swift:517
`GoalRecord` is a domain concept fused to persistence ON ITS OWN DECLARATION (the GRDB `PersistableRecord`/`FetchableRecord` conformance), with no separate pure-domain layer — the domain entity IS the ORM row (active-record). A schema/persistence change ripples straight into the domain type. Keep the ORM/table mapping on a boundary adapter and the domain entity framework-free (or map row↔domain through a repository). — GoalRecord.swift:4
`MemoryRecord` is a domain concept fused to persistence ON ITS OWN DECLARATION (the GRDB `PersistableRecord`/`FetchableRecord` conformance), with no separate pure-domain layer — the domain entity IS the ORM row (active-record). A schema/persistence change ripples straight into the domain type. Keep the ORM/table mapping on a boundary adapter and the domain entity framework-free (or map row↔domain through a repository). — MemoryModels.swift:8
`ObservationRecord` is a domain concept fused to persistence ON ITS OWN DECLARATION (the GRDB `PersistableRecord`/`FetchableRecord` conformance), with no separate pure-domain layer — the domain entity IS the ORM row (active-record). A schema/persistence change ripples straight into the domain type. Keep the ORM/table mapping on a boundary adapter and the domain entity framework-free (or map row↔domain through a repository). — ObservationRecord.swift:6
`ProactiveExtractionRecord` is a domain concept fused to persistence ON ITS OWN DECLARATION (the GRDB `PersistableRecord`/`FetchableRecord` conformance), with no separate pure-domain layer — the domain entity IS the ORM row (active-record). A schema/persistence change ripples straight into the domain type. Keep the ORM/table mapping on a boundary adapter and the domain entity framework-free (or map row↔domain through a repository). — ProactiveModels.swift:16
`FocusSessionRecord` is a domain concept fused to persistence ON ITS OWN DECLARATION (the GRDB `PersistableRecord`/`FetchableRecord` conformance), with no separate pure-domain layer — the domain entity IS the ORM row (active-record). A schema/persistence change ripples straight into the domain type. Keep the ORM/table mapping on a boundary adapter and the domain entity framework-free (or map row↔domain through a repository). — ProactiveModels.swift:92
`TaskDedupLogRecord` is a domain concept fused to persistence ON ITS OWN DECLARATION (the GRDB `PersistableRecord`/`FetchableRecord` conformance), with no separate pure-domain layer — the domain entity IS the ORM row (active-record). A schema/persistence change ripples straight into the domain type. Keep the ORM/table mapping on a boundary adapter and the domain entity framework-free (or map row↔domain through a repository). — ProactiveModels.swift:163
`Screenshot` is a domain concept fused to persistence ON ITS OWN DECLARATION (the GRDB `PersistableRecord`/`FetchableRecord` conformance), with no separate pure-domain layer — the domain entity IS the ORM row (active-record). A schema/persistence change ripples straight into the domain type. Keep the ORM/table mapping on a boundary adapter and the domain entity framework-free (or map row↔domain through a repository). — RewindModels.swift:7
`TaskChatMessageRecord` is a domain concept fused to persistence ON ITS OWN DECLARATION (the GRDB `PersistableRecord`/`FetchableRecord` conformance), with no separate pure-domain layer — the domain entity IS the ORM row (active-record). A schema/persistence change ripples straight into the domain type. Keep the ORM/table mapping on a boundary adapter and the domain entity framework-free (or map row↔domain through a repository). — TaskChatMessageStorage.swift:9
`TranscriptionSessionRecord` is a domain concept fused to persistence ON ITS OWN DECLARATION (the GRDB `PersistableRecord`/`FetchableRecord` conformance), with no separate pure-domain layer — the domain entity IS the ORM row (active-record). A schema/persistence change ripples straight into the domain type. Keep the ORM/table mapping on a boundary adapter and the domain entity framework-free (or map row↔domain through a repository). — TranscriptionModels.swift:49
`TranscriptionSegmentRecord` is a domain concept fused to persistence ON ITS OWN DECLARATION (the GRDB `PersistableRecord`/`FetchableRecord` conformance), with no separate pure-domain layer — the domain entity IS the ORM row (active-record). A schema/persistence change ripples straight into the domain type. Keep the ORM/table mapping on a boundary adapter and the domain entity framework-free (or map row↔domain through a repository). — TranscriptionModels.swift:233
What to do
Invert domain→infrastructure dependencies: declare interfaces in the domain, implement them in infrastructure (Dependency Inversion).
Other · Domain Modelling — Whether clusters of primitives that travel together (a missing value object) are extracted — a low-weight suggestion, LLM-confirmed when configured.
Method: Roslyn (DDD-gated): primitive parameter clusters recurring three or more times across signatures extracted, then confirmed by language model when configured. Advisory, low-weight.
Maturity · Maturity — Whether the repo and its projects have a README, and whether it's substantive and current.
Method: Filesystem scan: README presence, word count, and headings for depth; git history for staleness. Exhaustive across root and project dirs, deterministic.
What to do
Add a 'Testing' section to the root README — how to run the test suite.
Add a README to the 16 of 34 project(s) that lack one — worth up to 0.9 pts.
Maturity · Maturity — Whether the README actually describes the code that exists (LLM-judged, advisory).
Method: Judged by language model at low temperature: README accuracy versus actual projects, within a disclosed tolerance. Advisory, not a measured number.
Readiness · Readiness — Whether SAST, secret/dependency scanning and performance benchmarking are wired in (presence, not runtime).
Method: Filesystem scan: SAST configuration, dependency-update automation, secret scanning, and a benchmark harness or benchmark step — in this repository's own ecosystem. Exhaustive, deterministic.
What to do
Enable Dependabot/Renovate or a dependency-review gate.
Add gitleaks/trufflehog in CI to block PRs that introduce committed secrets.
Readiness · Readiness — Whether releases are automated and safely reversible (probes, rolling updates, approval gates) — from manifests/pipeline files, not the live environment.
Method: Filesystem scan: deployment manifests/IaC (K8s YAML, Helm, Terraform) for rolling updates, probes, approval gates, migration hooks. Exhaustive, deterministic.
Do you agree with this assessment?
P5 · DR & Backup7.0 / 10Strong✓ Tool-verified
Readiness · Readiness — Whether disaster recovery is planned and codified — backups, geo-recovery, RTO/RPO, persistence guarantees — from IaC + container manifests + docs, never the live cloud.
Method: Filesystem scan: disaster recovery, backup, geo-recovery, RTO/RPO, persistence guarantees from IaC, manifests, and docs. Exhaustive, deterministic, never a live environment.
What to do
Enable purge protection / soft-delete (and prevent_destroy on critical resources) so data stores can't be lost to an accidental or malicious delete.
Readiness · Readiness — Whether releases are traceable — a maintained changelog and explicit version stamping.
Method: Filesystem scan: changelog file presence and version tags in csproj or git tags. Exhaustive, deterministic.
Do you agree with this assessment?
R1 · Type Safety9.3 / 10Exemplary✓ Tool-verified
React / JS · Code Health — How much of the frontend is typed TypeScript vs untyped JavaScript.
Method: Frontend file inventory: the share of typed TypeScript vs untyped JavaScript across the source tree. Deterministic, exhaustive over frontend files.
What to do
Migrate the remaining .js/.jsx files to TypeScript.
React / JS · Code Health — Copy-pasted token-identical blocks across the frontend (the D4 clone algorithm over JS/TS tokens, D-386).
Method: Copy-pasted token-identical blocks across the frontend (the D4 clone algorithm run over JS/TS tokens). Deterministic.
desktop/windows/src/main/agentKernel/kernelCore.ts:374 · desktop/windows/src/main/agentKernel/kernelCore.ts:558 — all 2 copies are in the same file, and the CITED SPAN is not a self-contained block — it runs from inside one construct into the next (the tail of a branch plus the head of the following one, a run of switch arms, the end of a declaration plus the list that follows it) rather than covering a whole unit. So do not lift these lines literally: no call can be substituted for a half-open construct. Extract the enclosing repeated UNIT instead — the whole function, component or branch these lines sit in — and where the repetition IS the construct (a run of switch arms, a stack of near-identical declarations) replace it with one table or registry looked up by key rather than a helper each arm calls. The copies still drift apart the first time only one of them is edited, which is why this is reported. — kernelCore.ts:374
desktop/macos/agent/src/adapters/acp.ts:62 · desktop/macos/agent/src/adapters/local-subprocess.ts:117 — the 2 copies sit in sibling files in one directory, so check first whether one of them (or an existing module there) already owns this behaviour and the others should call it; otherwise extract it into one module in that directory and have each site call it. — acp.ts:62
desktop/macos/agent/src/runtime/tool-invocation-ledger.ts:182 · desktop/macos/agent/src/runtime/tool-invocation-ledger.ts:223 — all 2 copies are in the same file, and what repeats is a LIST OF ENTRIES rather than behaviour — the same names written out more than once. Extract them into one shared, exported constant and spread that constant into each site, rather than into a function the sites call: a list like this often lives in declarative metadata (a decorator's options object, a static configuration table) that a build step must be able to read statically, where a function call is not allowed. Adding an entry to one copy and not the other is the failure this prevents. — tool-invocation-ledger.ts:182
desktop/windows/scripts/agent-voice-gauntlet.mjs:116 · desktop/windows/scripts/run-vad-playback.mjs:117 · desktop/windows/scripts/run-voice-loop-check.mjs:114 — the 3 copies sit in sibling files in one directory, so check first whether one of them (or an existing module there) already owns this behaviour and the others should call it; otherwise extract it into one module in that directory and have each site call it. — agent-voice-gauntlet.mjs:116
desktop/windows/scripts/run-vad-playback.mjs:118 · desktop/windows/scripts/run-voice-loop-check.mjs:115 — the 2 copies are spread across 2 files, and the CITED SPAN is not a self-contained block — it runs from inside one construct into the next (the tail of a branch plus the head of the following one, a run of switch arms, the end of a declaration plus the list that follows it) rather than covering a whole unit. So do not lift these lines literally: no call can be substituted for a half-open construct. Extract the enclosing repeated UNIT instead — the whole function, component or branch these lines sit in — and where the repetition IS the construct (a run of switch arms, a stack of near-identical declarations) replace it with one table or registry looked up by key rather than a helper each arm calls. The copies still drift apart the first time only one of them is edited, which is why this is reported. — run-vad-playback.mjs:118
desktop/macos/agent/src/omi-tools-stdio.ts:328 · desktop/macos/agent/src/omi-tools-stdio.ts:344 — all 2 copies are in the same file, and the CITED SPAN is not a self-contained block — it runs from inside one construct into the next (the tail of a branch plus the head of the following one, a run of switch arms, the end of a declaration plus the list that follows it) rather than covering a whole unit. So do not lift these lines literally: no call can be substituted for a half-open construct. Extract the enclosing repeated UNIT instead — the whole function, component or branch these lines sit in — and where the repetition IS the construct (a run of switch arms, a stack of near-identical declarations) replace it with one table or registry looked up by key rather than a helper each arm calls. The copies still drift apart the first time only one of them is edited, which is why this is reported. — omi-tools-stdio.ts:328
desktop/macos/agent/src/runtime/conversation-journal.ts:2802 · desktop/macos/agent/src/runtime/conversation-journal.ts:3164 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. — conversation-journal.ts:2802
desktop/macos/agent/src/runtime/direct-control-execution.ts:9 · desktop/macos/agent/src/runtime/kernel-types.ts:154 — the 2 copies are spread across 2 files, and the CITED SPAN is not a self-contained block — it runs from inside one construct into the next (the tail of a branch plus the head of the following one, a run of switch arms, the end of a declaration plus the list that follows it) rather than covering a whole unit. So do not lift these lines literally: no call can be substituted for a half-open construct. Extract the enclosing repeated UNIT instead — the whole function, component or branch these lines sit in — and where the repetition IS the construct (a run of switch arms, a stack of near-identical declarations) replace it with one table or registry looked up by key rather than a helper each arm calls. The copies still drift apart the first time only one of them is edited, which is why this is reported. — direct-control-execution.ts:9
What to do
Extract the duplicated blocks into shared functions/components.
React / JS · Architecture — Conformance to the detected frontend architecture layout (feature-sliced / layered src) plus cross-package deep-import rules (D-386).
Method: Conformance to the detected frontend layout (feature-sliced / layered src) plus cross-package deep-import rules, over the module graph. Deterministic.
desktop/macos/agent/src/adapters/pi-mono.ts:211 reaches into another package with a relative path (../../../pi-mono-extension/index.ts) — import the package by name instead. — pi-mono.ts:211
desktop/macos/pi-mono-extension/index.ts:35 reaches into another package with a relative path (../agent/dist/runtime/node-tools.js) — import the package by name instead. — index.ts:35
desktop/macos/pi-mono-extension/index.ts:36 reaches into another package with a relative path (../agent/dist/runtime/omi-tool-manifest.js) — import the package by name instead. — index.ts:36
desktop/windows/src/main/codingAgent/pi-mono-extension/index.ts:52 reaches into another package with a relative path (../../agentKernel/omiToolManifest) — import the package by name instead. — index.ts:52
What to do
Fix the listed violations: import through public entries (package exports / slice index), never reach into another layer or package's internals.
React / JS · Code Health — Per-function cyclomatic/cognitive complexity from the token-level function scanner (D-386) — real branching, not a regex heuristic.
Method: Per-function cyclomatic/cognitive complexity from a token-level function scanner (real branching, not a regex heuristic), computed over every frontend function. Deterministic.
Branch-heavy code is where defects cluster — extract decisions into smaller functions. (×8) — index.ts:1893, control-tools.ts:808, voiceTurnMachine.ts:641, …
What to do
Break down the listed branch-heavy functions; aim P95 cyclomatic ≤ 5.
Do you agree with this assessment?
R3 · Large Files6.9 / 10Adequate✓ Tool-verified
React / JS · Code Health — How many source files exceed the large-file threshold.
Method: Components/modules exceeding the large-file threshold, counted exhaustively across the frontend source tree. Deterministic.
122 file(s) over 400 lines (counted as significant lines — blank lines excluded — over production source only, tests excluded), largest first: desktop/windows/src/renderer/src/lib/omiApi.generated.ts (15166), desktop/macos/agent/src/runtime/conversation-journal.ts (4147), desktop/macos/agent/src/index.ts (3815), desktop/macos/agent/src/runtime/sqlite-store.ts (3657), web/app/src/components/settings/SettingsPage.tsx (3476), desktop/macos/agent/src/runtime/control-tools.ts (2870) (+116 more).
What to do
Split each oversized file along the responsibilities already in it, into smaller focused modules in the same package.
Do you agree with this assessment?
R4 · Test Coverage6.3 / 10Adequate✓ Tool-verified
React / JS · Readiness — Static test reachability (D-386): the share of production files reachable from any test via the import graph — measured without running anything.
Method: Static test reachability: the share of production files reachable from any test via the import graph — measured without running anything. Deterministic.
No test imports this module directly or transitively. Import reachability cannot see a test that executes a file by path instead of importing it, nor one that drives it through a running browser by navigating to a URL — if neither does, no test reaches this one. (×8) — SettingsPage.tsx, page.tsx, api.ts, …
What to do
Add tests that import the unreached modules (directly or through their public entry).
React / JS · Readiness — How outdated the npm dependencies are (a maturity signal). JS/npm CVEs are scored separately in D33 (JS/npm Dependency Vulnerabilities).
Method: npm dependency staleness from manifest/registry metadata (a maturity signal; JS/npm CVEs are scored separately in D33). Deterministic.
What to do
Bump outdated dependencies to current versions to limit upgrade debt.
Do you agree with this assessment?
R6 · Tooling10.0 / 10Exemplary✓ Tool-verified
React / JS · Readiness — Whether the project wires up test, lint and typecheck — detected from each package.json script's COMMAND (eslint / tsc / vitest / jest / playwright), not just its name, and corroborated against CI-workflow invocations so a tool run only in CI still counts.
Method: package.json scanned for test/lint/typecheck script wiring. Deterministic presence check.
Do you agree with this assessment?
R7 · Dead Code6.8 / 10Adequate✓ Tool-verified
React / JS · Code Health — Files unreachable from every application/tooling/test entry point, and exports nothing imports (module-graph reachability, D-386).
Method: Dead code: files unreachable from every application/tooling/test entry point plus exports nothing imports, via module-graph reachability. Deterministic, exhaustive over the import graph.
Unreachable from the 355 application, 47 tooling and 838 test entry point(s) detected in this repo. Gate removals on your build/type-check — an undetected custom entry would make these reachable.
no import path from any entry point (355 application, 47 tooling, 838 test roots considered) (×7) — agent-tool-gauntlet.mjs, index.ts, agent-voice-gauntlet.mjs, …
What to do
Delete the dead files and unused exports — every line is maintenance cost and rebuild-estimate inflation with zero runtime value.
React / JS · Readiness — npm dependency truthfulness (D-386): unused dependencies, imports not declared anywhere, and type-/test-only packages shipped as production deps.
Method: npm dependency truthfulness: unused dependencies, imports declared nowhere, and type-/test-only packages shipped as production deps — from the manifest + import graph. Deterministic.
Imported but not declared in any reachable package.json — installs work only by hoisting accident. — ablation.mjs:16
Declared in the root package.json but never imported anywhere in that package or its workspace members — dead weight and attack surface. Verify against build tooling before removing. (×2)
Declared in web/admin/package.json but never imported anywhere in that package or its workspace members — dead weight and attack surface. Verify against build tooling before removing.
What to do
Remove unused dependencies, declare unlisted imports explicitly, and demote type-/test-only packages to devDependencies.
Do you agree with this assessment?
R9 · Circular Imports4.8 / 10Weak✓ Tool-verified
React / JS · Architecture — Import cycles in the module graph (D-386) — files that can only be understood and changed together.
Method: Import cycles in the module graph, detected exhaustively over JS/TS imports (the same cycle detection as the .NET coupling dimension). Deterministic.
Break each cycle by extracting the shared piece into a module both sides can import.
Do you agree with this assessment?
WCAG coverage — what static analysis assessed
Statically assessed 15 of 55 WCAG 2.2 Level A/AA success criteria (27%; ≈30% of the 50 WCAG 2.1 AA criteria for EN 301 549). The other 40 require runtime or manual evaluation. Partial signal only (a clean result is necessary, not sufficient; static analysis fully verifies none). This is accessibility readiness, not a conformance claim — a WCAG conformance claim requires manual evaluation (WCAG-EM 1.0).
The score is the rank-weighted fold of these lenses (worst-heaviest), each including its meta-dimensions; a lens with a Critical contributor is capped at Fair (its band reads "gated by …") and is never the strongest area however high its average.
Capped at Fair by a Critical contributor — resolve it before relying on this lens.
Not included — 57 check(s) not relevant to this codebase
These checks had nothing to measure here (no tests, no git history, the codebase is small, or the architecture style doesn't apply), so they're omitted above rather than scored low.
AX1 Captive dependencies — no DI registrations detected
AX10 Code composition — not assessed — code composition is computed by ROLE over a document set that was not loaded for this repository, because the repository is written in a language this check does not yet model, or because its projects failed to load. This is a gap in the analyzer, not a finding about this repository
AX2 Stateful singletons — no singleton implementations detected
AX5 Architecture & structure — not assessed — architecture style/structure is computed from a project graph (projects, types, module namespaces) that was not loaded for this repository, because the repository is written in a language this check does not yet model, or because its projects failed to load. This is a gap in the analyzer, not a finding about this repository
AX6 Interface segregation — not assessed — interface segregation is computed over a type surface that was not loaded for this repository, because the repository is written in a language this check does not yet model, or because its projects failed to load. This is a gap in the analyzer, not a finding about this repository
AX7 Slice cohesion — not applicable — not a vertical-slice architecture
AX8 Test isolation — no test/production split to check
AXB2 Runtime readiness — Advisory — this card reports evidence and never carries a score, so there is nothing missing here.
C1 Data Protection — Not assessed: these personal data controls are read from a source model (declarative annotations, request middleware, entity/column names, guard methods) that was not loaded for this repository — because the repository is written in a language this check does not yet model, or because its projects failed to load. Absence of an idiom this check recognises is NOT evidence that this repository lacks personal data controls: it may implement them entirely in its own ecosystem. This is a gap in the analyzer's language coverage, not a finding about this repository.
C2 Access Controls — Not assessed: these authorization controls are read from a source model (declarative annotations, request middleware, entity/column names, guard methods) that was not loaded for this repository — because the repository is written in a language this check does not yet model, or because its projects failed to load. Absence of an idiom this check recognises is NOT evidence that this repository lacks authorization controls: it may implement them entirely in its own ecosystem. This is a gap in the analyzer's language coverage, not a finding about this repository.
C3 Audit Trail — Not assessed: these audit controls are read from a source model (declarative annotations, request middleware, entity/column names, guard methods) that was not loaded for this repository — because the repository is written in a language this check does not yet model, or because its projects failed to load. Absence of an idiom this check recognises is NOT evidence that this repository lacks audit controls: it may implement them entirely in its own ecosystem. This is a gap in the analyzer's language coverage, not a finding about this repository.
C4 Data Retention — Not assessed: these retention controls are read from a source model (declarative annotations, request middleware, entity/column names, guard methods) that was not loaded for this repository — because the repository is written in a language this check does not yet model, or because its projects failed to load. Absence of an idiom this check recognises is NOT evidence that this repository lacks retention controls: it may implement them entirely in its own ecosystem. This is a gap in the analyzer's language coverage, not a finding about this repository.
C5 Data-Subject Rights — Not assessed: these data-subject rights controls are read from a source model (declarative annotations, request middleware, entity/column names, guard methods) that was not loaded for this repository — because the repository is written in a language this check does not yet model, or because its projects failed to load. Absence of an idiom this check recognises is NOT evidence that this repository lacks data-subject rights controls: it may implement them entirely in its own ecosystem. This is a gap in the analyzer's language coverage, not a finding about this repository.
D1 Cyclomatic Complexity — Most of this repository's production source (.cs, .dart, .kt, .rs, .swift, .ts, .tsx) had no cyclomatic complexity computed for it, so cyclomatic complexity was not measured — whatever else this pass did read is not this repository's complexity. Not scored: no method bodies were exposed for those file kinds by any language model this pass could load. This is a gap in the analysis run, not a finding about this repository.
D10 Test Quality — ~437620 lines of test source are present (.dart, .py, .swift, .ts, .tsx, .mjs, .js, .rs) but the test-quality collector reads C# only, so skipped/assertion-free tests couldn't be counted. Not scored — this is a gap in the analyzer, not a finding about this repository.
D11 Test Reliability — Test reliability not included
D12 Dependency Hygiene — Dependency hygiene not measured — dependency manifest found but not parsed for hygiene
D17 Explicit Debt — the C# workspace loaded 0 projects, so explicit-debt density could not be measured
D18 Solution Shape — D18 scores the shape of a .NET solution, but this repository's production source is mostly .dart, .go, .kt, .py, .rb, .rs, .swift, .ts, .tsx — the .NET project files present are an immaterial minority — so the dimension does not apply.
D2 Cognitive Complexity — Most of this repository's production source (.cs, .dart, .kt, .rs, .swift, .ts, .tsx) had no cognitive complexity computed for it, so cognitive complexity was not measured — whatever else this pass did read is not this repository's complexity. Not scored: no method bodies were exposed for those file kinds by any language model this pass could load. This is a gap in the analysis run, not a finding about this repository.
D22 Internal API Consistency — No exposed public API
D23 Boundary Type-Coupling — Production source is present (.cs, .dart, .go, .kt, .py, .rb, .rs, .swift, .ts, .tsx) but bounded contexts are resolved over the C#/VB project set, which exposed none, so context scope could not be assessed. Not scored — this is a gap in the analyzer, not a verdict about this repository. Declaring the codebase's bounded contexts (≥2) would let cross-boundary type coupling be assessed — see the recommendation on this dimension for where. Declare them in `.codehealth/config.yaml` at the repository root (create it if absent), mapping each context name to the module-path or namespace prefixes that belong to it — e.g. `architecture:` → `contexts:` → `Billing: ["src/billing", "Acme.Billing"]`, `Catalog: ["src/catalog", "Acme.Catalog"]`.
D24 Comment Value — No inline comments to assess — comment value is not applicable here.
D25 ADR Conformance — none of 2 ADRs are conformance-checkable — unverifiable.
D26 Project Cohesion — Project cohesion is assessed over the .NET project set; this target exposed no projects, so project size and spread could not be assessed. Not scored — this is a gap in the analyzer's reach, not a verdict about this repository.
D27 Navigability — No calls could be sampled, so navigability was not assessed — tracing effort is measured over resolved call sites and this target exposed none. Not scored — this is a gap in the analyzer's reach, not a verdict about this repository.
D3 God Classes — Most of this repository's production source (.cs, .dart, .kt, .rs, .swift, .ts, .tsx) was not read by god-class detection, so class size was not assessed for the languages that are the product — whatever else this pass did read is not this repository's class size. Not scored — this is a gap in the analyzer, not a verdict about this repository.
D30 Dependency Vulnerabilities — Not scored — no dependency manifest in a supported ecosystem was read for this repository. A gap in the analyzer's language coverage, NOT a finding that the repository is free of vulnerable dependencies (a Swift Package.swift/Package.resolved, package.json, a Python pyproject.toml/requirements.txt (pip/uv/Poetry), a Cargo manifest and a Dart pubspec.yaml — not scanned yet) — where an OSV-supported manifest exists, dependency vulnerabilities for this repository are reported under D38 instead.
D39 IL Efficiency — The target did not build, so no IL was available to measure.
D6 Cohesion (LCOM4) — Cohesion (LCOM4) is measured over a C#/VB class graph, and this repository's production source is .dart, .go, .kt, .py, .rb, .rs, .swift, .ts, .tsx, which this pass does not read — so no class could be assessed. Not scored — this is a gap in the analyzer, not a finding about this repository.
D8 Code Coverage — Coverage not included — suite not readable by the collector
D9 Test Distribution — Test source is present (.dart, .py, .swift, .ts, .tsx, .mjs, .js, .rs) but the test-pyramid classifier reads C# only, so its unit/integration/BDD/E2E split couldn't be counted. Not scored — this is a gap in the analyzer, not a finding about this repository.
DM2 Strongly-typed ids — no id-bearing domain types detected — strongly-typed-id adoption not assessable
DM3 Integration-event coupling — no integration events detected — coupling check not applicable
DM7 Repository granularity — no repository abstraction detected (e.g. uses a document session)
ED2 Event/command shape — not scored — deciding whether a command has more than one competing handler requires resolving the call graph, and this analysis resolves a call's owner only where the receiver's type is written down in the source. Reported as guidance rather than measured
ED2 Event/command shape — not scored — deciding whether a command has more than one competing handler requires resolving the call graph, which this analysis does not build from source alone for this language. Reported as guidance rather than measured
ED2 Event/command shape — not scored — deciding whether a command has more than one competing handler requires resolving the call graph, and a call made through an inferred or generic receiver has no resolvable owner in the source. Reported as guidance rather than measured
ED5 Idempotency — no mutating command handlers or message consumers detected — idempotency check not applicable
ES1 Event Sourcing — not scored — this repository shows none of the 3 signals this check looks for
GD1 Unfinished & placeholder code — no source files
IC1 Incompleteness & stubs — not analysed — these correctness checks read a source model that was not loaded for this repository, because the repository is written in a language this check does not yet model, or because its projects failed to load. This is a gap in the analyzer, not a finding about this repository
P12 CI test-gate honesty — Reported, not scored — and nothing was matched here. The coverage check applies to any stack, but the checks for excluded tests, skipped tests and sleep-based synchronisation currently recognise only some ecosystems' test-runner idioms, so on a repository built with another stack the zeros below mean 'not checked', not 'clean'.
P2 Observability — Observability was not assessed: this check reads a source model that does not carry this repository's product — because the repository is written in a language this check does not yet model, or because its projects failed to load. Absence of a logging idiom this check recognises is NOT evidence that this repo lacks structured logging (it may log through its own ecosystem's logger). This is a gap in the analyzer, not a finding about this repository.
P7 Outbound HTTP resilience — not measured — the application kind could not be determined for this repo
P8 Schema migrations — not assessed — schema-migration practice is read from a source model that was not loaded for this repository, because the repository is written in a language this check does not yet model or because its projects failed to load. This is a gap in the analyzer, not a finding about this repository
P9 Domain vs controller coverage — no coverage report found on disk — produce a coverage report in a standard format (`coverage run -m pytest` then `coverage xml`) into the repo working tree before the scan — a CI step is the usual place, since the artefact is commonly gitignored, or wire coverage collection into CI, to enable this cross-layer check
PF1 Benchmark discipline — Performance was not assessed: this lens reads a source model that was not loaded for this repository, because the repository is written in a language this lens does not yet model or because its projects failed to load. This is a gap in the analyzer, not a finding about this repository — in particular it is NOT a statement that this repo is unpackaged or performance-careless.
PF2 Allocation hygiene — Performance was not assessed: this lens reads a source model that was not loaded for this repository, because the repository is written in a language this lens does not yet model or because its projects failed to load. This is a gap in the analyzer, not a finding about this repository — in particular it is NOT a statement that this repo is unpackaged or performance-careless.
PF3 Async & latency hygiene — Performance was not assessed: this lens reads a source model that was not loaded for this repository, because the repository is written in a language this lens does not yet model or because its projects failed to load. This is a gap in the analyzer, not a finding about this repository — in particular it is NOT a statement that this repo is unpackaged or performance-careless.
S1 Web-Security Posture — Not assessed: these web-security controls are read from a source model (declarative annotations, request middleware, entity/column names, guard methods) that was not loaded for this repository — because the repository is written in a language this check does not yet model, or because its projects failed to load. Absence of an idiom this check recognises is NOT evidence that this repository lacks web-security controls: it may implement them entirely in its own ecosystem. This is a gap in the analyzer's language coverage, not a finding about this repository.
SC1 Supply-chain hygiene — Advisory — this card reports evidence and never carries a score, so there is nothing missing here.
X1 Async correctness — not analysed — these correctness checks read a source model that was not loaded for this repository, because the repository is written in a language this check does not yet model, or because its projects failed to load. This is a gap in the analyzer, not a finding about this repository
X2 Cancellation propagation — not analysed — these correctness checks read a source model that was not loaded for this repository, because the repository is written in a language this check does not yet model, or because its projects failed to load. This is a gap in the analyzer, not a finding about this repository
X3 Exception handling — not analysed — these correctness checks read a source model that was not loaded for this repository, because the repository is written in a language this check does not yet model, or because its projects failed to load. This is a gap in the analyzer, not a finding about this repository
X4 Structured logging — not analysed — these correctness checks read a source model that was not loaded for this repository, because the repository is written in a language this check does not yet model, or because its projects failed to load. This is a gap in the analyzer, not a finding about this repository
X5 Nullable reference types — not analysed — these correctness checks read a source model that was not loaded for this repository, because the repository is written in a language this check does not yet model, or because its projects failed to load. This is a gap in the analyzer, not a finding about this repository
Appendix A — Findings (grouped)
The findings behind the scores, grouped by severity, then by dimension and kind. The high-severity issues are enumerated in full below; items per group are capped at 25 with any overflow stated explicitly per group, never silently truncated. The complete machine-readable list of every finding (all severities) is the companion findings.md in this report's bundle.
High: run-shell-injection .github/actions/deploy-backend-stack/action.yml:78— Using variable interpolation `${{...}}` with a workflow input in a `run:` step could allow an attacker to inject their own code into the runner. This would allow them to steal secrets and code. A workflow input is not bounded by this step and should be treated as untrusted. Instead, use an intermediate environment variable with `env:` to store the data and use the environment variable in the `run:` script. Reference it as a shell VARIABLE rather than a `${{ }}` interpolation, using your shell's own syntax (`"$ENVVAR"` in bash, `$env:ENVVAR` in PowerShell), so the value is passed as data and never re-expanded as code.
High: github-actions-mutable-action-tag .github/actions/deploy-backend-stack/action.yml:104— GitHub Actions step uses a mutable tag or branch reference. Tags and branch names can be silently repointed by the action owner, enabling supply-chain attacks — as seen in the trivy-action and kics-github-action compromises. Pin the reference to a full 40-character commit SHA instead, e.g. `uses: actions/checkout@<40-character SHA>`. This step references `actions/checkout@v7`; resolve the SHA it points at today with `gh api repos/actions/checkout/commits/v7 --jq .sha`.
High: github-actions-mutable-action-tag .github/actions/deploy-backend-stack/action.yml:131— GitHub Actions step uses a mutable tag or branch reference. Tags and branch names can be silently repointed by the action owner, enabling supply-chain attacks — as seen in the trivy-action and kics-github-action compromises. Pin the reference to a full 40-character commit SHA instead, e.g. `uses: actions/checkout@<40-character SHA>`. This step references `actions/checkout@v7`; resolve the SHA it points at today with `gh api repos/actions/checkout/commits/v7 --jq .sha`.
High: github-actions-mutable-action-tag .github/actions/deploy-backend-stack/action.yml:156— GitHub Actions step uses a mutable tag or branch reference. Tags and branch names can be silently repointed by the action owner, enabling supply-chain attacks — as seen in the trivy-action and kics-github-action compromises. Pin the reference to a full 40-character commit SHA instead, e.g. `uses: google-github-actions/auth@<40-character SHA>`. This step references `google-github-actions/auth@v3`; resolve the SHA it points at today with `gh api repos/google-github-actions/auth/commits/v3 --jq .sha`.
High: github-actions-mutable-action-tag .github/actions/deploy-backend-stack/action.yml:161— GitHub Actions step uses a mutable tag or branch reference. Tags and branch names can be silently repointed by the action owner, enabling supply-chain attacks — as seen in the trivy-action and kics-github-action compromises. Pin the reference to a full 40-character commit SHA instead, e.g. `uses: google-github-actions/setup-gcloud@<40-character SHA>`. This step references `google-github-actions/setup-gcloud@v3`; resolve the SHA it points at today with `gh api repos/google-github-actions/setup-gcloud/commits/v3 --jq .sha`.
High: run-shell-injection .github/actions/deploy-backend-stack/action.yml:167— Using variable interpolation `${{...}}` with a workflow input in a `run:` step could allow an attacker to inject their own code into the runner. This would allow them to steal secrets and code. A workflow input is not bounded by this step and should be treated as untrusted. Instead, use an intermediate environment variable with `env:` to store the data and use the environment variable in the `run:` script. Reference it as a shell VARIABLE rather than a `${{ }}` interpolation, using your shell's own syntax (`"$ENVVAR"` in bash, `$env:ENVVAR` in PowerShell), so the value is passed as data and never re-expanded as code.
High: github-actions-mutable-action-tag .github/actions/deploy-backend-stack/action.yml:179— GitHub Actions step uses a mutable tag or branch reference. Tags and branch names can be silently repointed by the action owner, enabling supply-chain attacks — as seen in the trivy-action and kics-github-action compromises. Pin the reference to a full 40-character commit SHA instead, e.g. `uses: docker/setup-buildx-action@<40-character SHA>`. This step references `docker/setup-buildx-action@v4`; resolve the SHA it points at today with `gh api repos/docker/setup-buildx-action/commits/v4 --jq .sha`.
High: run-shell-injection .github/actions/deploy-backend-stack/action.yml:197— Using variable interpolation `${{...}}` with a workflow input in a `run:` step could allow an attacker to inject their own code into the runner. This would allow them to steal secrets and code. A workflow input is not bounded by this step and should be treated as untrusted. Instead, use an intermediate environment variable with `env:` to store the data and use the environment variable in the `run:` script. Reference it as a shell VARIABLE rather than a `${{ }}` interpolation, using your shell's own syntax (`"$ENVVAR"` in bash, `$env:ENVVAR` in PowerShell), so the value is passed as data and never re-expanded as code.
High: run-shell-injection .github/actions/deploy-backend-stack/action.yml:226— Using variable interpolation `${{...}}` with a workflow input in a `run:` step could allow an attacker to inject their own code into the runner. This would allow them to steal secrets and code. A workflow input is not bounded by this step and should be treated as untrusted. Instead, use an intermediate environment variable with `env:` to store the data and use the environment variable in the `run:` script. Reference it as a shell VARIABLE rather than a `${{ }}` interpolation, using your shell's own syntax (`"$ENVVAR"` in bash, `$env:ENVVAR` in PowerShell), so the value is passed as data and never re-expanded as code.
High: github-actions-mutable-action-tag .github/actions/deploy-backend-stack/action.yml:239— GitHub Actions step uses a mutable tag or branch reference. Tags and branch names can be silently repointed by the action owner, enabling supply-chain attacks — as seen in the trivy-action and kics-github-action compromises. Pin the reference to a full 40-character commit SHA instead, e.g. `uses: google-github-actions/get-gke-credentials@<40-character SHA>`. This step references `google-github-actions/get-gke-credentials@v3`; resolve the SHA it points at today with `gh api repos/google-github-actions/get-gke-credentials/commits/v3 --jq .sha`.
High: run-shell-injection .github/actions/deploy-backend-stack/action.yml:252— Using variable interpolation `${{...}}` with a workflow input in a `run:` step could allow an attacker to inject their own code into the runner. This would allow them to steal secrets and code. A workflow input is not bounded by this step and should be treated as untrusted. Instead, use an intermediate environment variable with `env:` to store the data and use the environment variable in the `run:` script. Reference it as a shell VARIABLE rather than a `${{ }}` interpolation, using your shell's own syntax (`"$ENVVAR"` in bash, `$env:ENVVAR` in PowerShell), so the value is passed as data and never re-expanded as code.
High: run-shell-injection .github/actions/deploy-backend-stack/action.yml:261— Using variable interpolation `${{...}}` with a workflow input in a `run:` step could allow an attacker to inject their own code into the runner. This would allow them to steal secrets and code. A workflow input is not bounded by this step and should be treated as untrusted. Instead, use an intermediate environment variable with `env:` to store the data and use the environment variable in the `run:` script. Reference it as a shell VARIABLE rather than a `${{ }}` interpolation, using your shell's own syntax (`"$ENVVAR"` in bash, `$env:ENVVAR` in PowerShell), so the value is passed as data and never re-expanded as code.
High: run-shell-injection .github/actions/deploy-backend-stack/action.yml:281— Using variable interpolation `${{...}}` with a workflow input in a `run:` step could allow an attacker to inject their own code into the runner. This would allow them to steal secrets and code. A workflow input is not bounded by this step and should be treated as untrusted. Instead, use an intermediate environment variable with `env:` to store the data and use the environment variable in the `run:` script. Reference it as a shell VARIABLE rather than a `${{ }}` interpolation, using your shell's own syntax (`"$ENVVAR"` in bash, `$env:ENVVAR` in PowerShell), so the value is passed as data and never re-expanded as code.
High: github-actions-mutable-action-tag .github/actions/deploy-backend-stack/action.yml:290— GitHub Actions step uses a mutable tag or branch reference. Tags and branch names can be silently repointed by the action owner, enabling supply-chain attacks — as seen in the trivy-action and kics-github-action compromises. Pin the reference to a full 40-character commit SHA instead, e.g. `uses: docker/build-push-action@<40-character SHA>`. This step references `docker/build-push-action@v7`; resolve the SHA it points at today with `gh api repos/docker/build-push-action/commits/v7 --jq .sha`.
High: run-shell-injection .github/actions/deploy-backend-stack/action.yml:304— Using variable interpolation `${{...}}` with a step output / a workflow input in a `run:` step could allow an attacker to inject their own code into the runner. This would allow them to steal secrets and code. A step output / a workflow input is not bounded by this step and should be treated as untrusted. Instead, use an intermediate environment variable with `env:` to store the data and use the environment variable in the `run:` script. Reference it as a shell VARIABLE rather than a `${{ }}` interpolation, using your shell's own syntax (`"$ENVVAR"` in bash, `$env:ENVVAR` in PowerShell), so the value is passed as data and never re-expanded as code.
High: run-shell-injection .github/actions/deploy-backend-stack/action.yml:310— Using variable interpolation `${{...}}` with a step output / a workflow input in a `run:` step could allow an attacker to inject their own code into the runner. This would allow them to steal secrets and code. A step output / a workflow input is not bounded by this step and should be treated as untrusted. Instead, use an intermediate environment variable with `env:` to store the data and use the environment variable in the `run:` script. Reference it as a shell VARIABLE rather than a `${{ }}` interpolation, using your shell's own syntax (`"$ENVVAR"` in bash, `$env:ENVVAR` in PowerShell), so the value is passed as data and never re-expanded as code.
High: github-actions-mutable-action-tag .github/actions/deploy-backend-stack/action.yml:316— GitHub Actions step uses a mutable tag or branch reference. Tags and branch names can be silently repointed by the action owner, enabling supply-chain attacks — as seen in the trivy-action and kics-github-action compromises. Pin the reference to a full 40-character commit SHA instead, e.g. `uses: azure/setup-helm@<40-character SHA>`. This step references `azure/setup-helm@v5`; resolve the SHA it points at today with `gh api repos/azure/setup-helm/commits/v5 --jq .sha`.
High: run-shell-injection .github/actions/deploy-backend-stack/action.yml:321— Using variable interpolation `${{...}}` with a step output / a workflow input in a `run:` step could allow an attacker to inject their own code into the runner. This would allow them to steal secrets and code. A step output / a workflow input is not bounded by this step and should be treated as untrusted. Instead, use an intermediate environment variable with `env:` to store the data and use the environment variable in the `run:` script. Reference it as a shell VARIABLE rather than a `${{ }}` interpolation, using your shell's own syntax (`"$ENVVAR"` in bash, `$env:ENVVAR` in PowerShell), so the value is passed as data and never re-expanded as code.
High: run-shell-injection .github/actions/deploy-backend-stack/action.yml:338— Using variable interpolation `${{...}}` with a step output / a workflow input in a `run:` step could allow an attacker to inject their own code into the runner. This would allow them to steal secrets and code. A step output / a workflow input is not bounded by this step and should be treated as untrusted. Instead, use an intermediate environment variable with `env:` to store the data and use the environment variable in the `run:` script. Reference it as a shell VARIABLE rather than a `${{ }}` interpolation, using your shell's own syntax (`"$ENVVAR"` in bash, `$env:ENVVAR` in PowerShell), so the value is passed as data and never re-expanded as code.
High: run-shell-injection .github/actions/deploy-backend-stack/action.yml:351— Using variable interpolation `${{...}}` with a workflow input in a `run:` step could allow an attacker to inject their own code into the runner. This would allow them to steal secrets and code. A workflow input is not bounded by this step and should be treated as untrusted. Instead, use an intermediate environment variable with `env:` to store the data and use the environment variable in the `run:` script. Reference it as a shell VARIABLE rather than a `${{ }}` interpolation, using your shell's own syntax (`"$ENVVAR"` in bash, `$env:ENVVAR` in PowerShell), so the value is passed as data and never re-expanded as code.
High: run-shell-injection .github/actions/deploy-backend-stack/action.yml:361— Using variable interpolation `${{...}}` with a step output / a workflow input in a `run:` step could allow an attacker to inject their own code into the runner. This would allow them to steal secrets and code. A step output / a workflow input is not bounded by this step and should be treated as untrusted. Instead, use an intermediate environment variable with `env:` to store the data and use the environment variable in the `run:` script. Reference it as a shell VARIABLE rather than a `${{ }}` interpolation, using your shell's own syntax (`"$ENVVAR"` in bash, `$env:ENVVAR` in PowerShell), so the value is passed as data and never re-expanded as code.
High: run-shell-injection .github/actions/deploy-backend-stack/action.yml:379— Using variable interpolation `${{...}}` with a step output / a workflow input / an `env` value in a `run:` step could allow an attacker to inject their own code into the runner. This would allow them to steal secrets and code. A step output / a workflow input / an `env` value is not bounded by this step and should be treated as untrusted. Instead, use an intermediate environment variable with `env:` to store the data and use the environment variable in the `run:` script. Reference it as a shell VARIABLE rather than a `${{ }}` interpolation, using your shell's own syntax (`"$ENVVAR"` in bash, `$env:ENVVAR` in PowerShell), so the value is passed as data and never re-expanded as code.
High: run-shell-injection .github/actions/deploy-backend-stack/action.yml:428— Using variable interpolation `${{...}}` with a workflow input in a `run:` step could allow an attacker to inject their own code into the runner. This would allow them to steal secrets and code. A workflow input is not bounded by this step and should be treated as untrusted. Instead, use an intermediate environment variable with `env:` to store the data and use the environment variable in the `run:` script. Reference it as a shell VARIABLE rather than a `${{ }}` interpolation, using your shell's own syntax (`"$ENVVAR"` in bash, `$env:ENVVAR` in PowerShell), so the value is passed as data and never re-expanded as code.
High: run-shell-injection .github/actions/deploy-backend-stack/action.yml:435— Using variable interpolation `${{...}}` with a workflow input in a `run:` step could allow an attacker to inject their own code into the runner. This would allow them to steal secrets and code. A workflow input is not bounded by this step and should be treated as untrusted. Instead, use an intermediate environment variable with `env:` to store the data and use the environment variable in the `run:` script. Reference it as a shell VARIABLE rather than a `${{ }}` interpolation, using your shell's own syntax (`"$ENVVAR"` in bash, `$env:ENVVAR` in PowerShell), so the value is passed as data and never re-expanded as code.
High: run-shell-injection .github/actions/deploy-backend-stack/action.yml:448— Using variable interpolation `${{...}}` with a workflow input in a `run:` step could allow an attacker to inject their own code into the runner. This would allow them to steal secrets and code. A workflow input is not bounded by this step and should be treated as untrusted. Instead, use an intermediate environment variable with `env:` to store the data and use the environment variable in the `run:` script. Reference it as a shell VARIABLE rather than a `${{ }}` interpolation, using your shell's own syntax (`"$ENVVAR"` in bash, `$env:ENVVAR` in PowerShell), so the value is passed as data and never re-expanded as code.
D38 · OSV Dependency Vulnerabilities· High CVE · ×31
High CVE: [GHSA redacted] backend/diarizer/requirements.txt— aiohttp 3.13.4: [GHSA redacted] — upgrade to 3.14.3. This is 1 of 14 advisories with a published fix this scan raises against aiohttp 3.13.4, and their fixed versions do not agree — anything below 3.14.3 still leaves at least one of them open. Take this package to 3.14.3 or later: that is the floor for the package, not this row's target alone. This one row stands for the 14 advisories this scan raises against aiohttp 3.13.4: [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], PYSEC-2026-2106, PYSEC-2026-2112, PYSEC-2026-2113, PYSEC-2026-237.
High CVE: [GHSA redacted] plugins/omi-hive-app/requirements.txt— aiohttp 3.13.5: [GHSA redacted] — upgrade to 3.14.3. This is 1 of 14 advisories with a published fix this scan raises against aiohttp 3.13.5, and their fixed versions do not agree — anything below 3.14.3 still leaves at least one of them open. Take this package to 3.14.3 or later: that is the floor for the package, not this row's target alone. (in 4 dependency files) This one row stands for the 14 advisories this scan raises against aiohttp 3.13.5: [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], PYSEC-2026-2106, PYSEC-2026-2112, PYSEC-2026-2113, PYSEC-2026-237.
High CVE: [GHSA redacted] backend/modal/requirements.txt— aiohttp 3.14.1: [GHSA redacted] — aiohttp is not declared by the manifests that govern this lockfile: it is pulled in transitively here (the 3.13.5 this repo declares for it belongs to a separate install root and does not reach this one), so upgrade the dependency that requires it (or raise its floor in your own manifest and re-resolve (`uv lock --upgrade-package aiohttp`)). This is 1 of 3 advisories with a published fix this scan raises against aiohttp 3.14.1, and their fixed versions do not agree — anything below 3.14.3 still leaves at least one of them open. Take this package to 3.14.3 or later: that is the floor for the package, not this row's target alone. (in 2 dependency files: backend/modal/requirements.txt, backend/requirements.txt) This one row stands for the 3 advisories this scan raises against aiohttp 3.14.1: [GHSA redacted], [GHSA redacted], [GHSA redacted].
High CVE: [GHSA redacted] plugins/zapier/omi-zap/package-lock.json— brace-expansion 1.1.14: [GHSA redacted] — brace-expansion is not declared by the manifests that govern this lockfile: it is pulled in transitively here (the ~5.0.8 this repo declares for it belongs to a separate install root and does not reach this one), so upgrade the dependency that requires it (or pin brace-expansion to 1.1.16 with an `overrides` entry (`pnpm.overrides` for pnpm)). This is 1 of 3 advisories with a published fix this scan raises against brace-expansion 1.1.14, and their fixed versions do not agree — anything below 1.1.18 still leaves at least one of them open. Take this package to 1.1.18 or later: that is the floor for the package, not this row's target alone. This one row stands for the 3 advisories this scan raises against brace-expansion 1.1.14: [GHSA redacted], [GHSA redacted], [GHSA redacted].
High CVE: [GHSA redacted] desktop/windows/pnpm-lock.yaml— brace-expansion 1.1.16: [GHSA redacted] — brace-expansion is not declared by the manifests that govern this lockfile: it is pulled in transitively here (the ~5.0.8 this repo declares for it belongs to a separate install root and does not reach this one), so upgrade the dependency that requires it (or pin brace-expansion to 1.1.17 with an `overrides` entry (`pnpm.overrides` for pnpm)). This is 1 of 2 advisories with a published fix this scan raises against brace-expansion 1.1.16, and their fixed versions do not agree — anything below 1.1.18 still leaves at least one of them open. Take this package to 1.1.18 or later: that is the floor for the package, not this row's target alone. (in 4 dependency files) This one row stands for the 2 advisories this scan raises against brace-expansion 1.1.16: [GHSA redacted], [GHSA redacted].
High CVE: [GHSA redacted] desktop/windows/pnpm-lock.yaml— brace-expansion 2.1.2: [GHSA redacted] — brace-expansion is not declared by the manifests that govern this lockfile: it is pulled in transitively here (the ~5.0.8 this repo declares for it belongs to a separate install root and does not reach this one), so upgrade the dependency that requires it (or pin brace-expansion to 2.1.3 with an `overrides` entry (`pnpm.overrides` for pnpm)). This is 1 of 2 advisories with a published fix this scan raises against brace-expansion 2.1.2, and their fixed versions do not agree — anything below 2.1.4 still leaves at least one of them open. Take this package to 2.1.4 or later: that is the floor for the package, not this row's target alone. (in 4 dependency files) This one row stands for the 2 advisories this scan raises against brace-expansion 2.1.2: [GHSA redacted], [GHSA redacted].
High CVE: [GHSA redacted] desktop/macos/agent/package-lock.json— brace-expansion 5.0.7: [GHSA redacted] — this repo already declares brace-expansion ~5.0.8 directly, at or above the fixed 5.0.8; the vulnerable 5.0.7 is a second copy resolved for a dependency that requires an older range. Upgrade those dependents, or pin brace-expansion with an `overrides` entry (`pnpm.overrides` for pnpm) so only one copy resolves. This is 1 of 2 advisories with a published fix this scan raises against brace-expansion 5.0.7, and their fixed versions do not agree — anything below 5.0.9 still leaves at least one of them open. Take this package to 5.0.9 or later: that is the floor for the package, not this row's target alone. (in 4 dependency files) This one row stands for the 2 advisories this scan raises against brace-expansion 5.0.7: [GHSA redacted], [GHSA redacted].
High CVE: [GHSA redacted] desktop/windows/pnpm-lock.yaml— brace-expansion 5.0.8: [GHSA redacted] — brace-expansion is not declared by the manifests that govern this lockfile: it is pulled in transitively here (the ~5.0.8 this repo declares for it belongs to a separate install root and does not reach this one), so upgrade the dependency that requires it (or pin brace-expansion to 5.0.9 with an `overrides` entry (`pnpm.overrides` for pnpm)). (in 4 dependency files)
High CVE: [GHSA redacted] backend/modal/requirements.txt— cryptography 48.0.1: [GHSA redacted] — cryptography is not declared by the manifests that govern this lockfile: it is pulled in transitively here (the 48.0.0 this repo declares for it belongs to a separate install root and does not reach this one), so upgrade the dependency that requires it (or raise its floor in your own manifest and re-resolve (`uv lock --upgrade-package cryptography`)). This is 1 of 3 advisories with a published fix this scan raises against cryptography 48.0.1, and their fixed versions do not agree — anything below 50.0.0 still leaves at least one of them open. Take this package to 50.0.0 or later: that is the floor for the package, not this row's target alone. (in 4 dependency files) This one row stands for the 3 advisories this scan raises against cryptography 48.0.1: [GHSA redacted], [GHSA redacted], [GHSA redacted].
High CVE: [GHSA redacted] desktop/windows/pnpm-lock.yaml— fast-uri 3.1.4: [GHSA redacted] — fast-uri is not declared by the manifests that govern this lockfile: it is pulled in transitively here (the ~3.1.4 this repo declares for it belongs to a separate install root and does not reach this one), so upgrade the dependency that requires it (or pin fast-uri to 3.1.5 with an `overrides` entry (`pnpm.overrides` for pnpm)). (in 3 dependency files: desktop/windows/pnpm-lock.yaml, package-lock.json, sdks/react-native/package-lock.json)
High CVE: [GHSA redacted] plugins/zapier/omi-zap/package-lock.json— form-data 4.0.5: [GHSA redacted] — this repo declares form-data ^4.0.4, a range that ALREADY admits the fixed 4.0.6, so there is no manifest edit to make here. Re-resolve the lock so form-data moves onto 4.0.6 or later; if the flagged 4.0.5 comes back, a dependency is pinning it — upgrade that dependent, or pin form-data with an `overrides` entry (`pnpm.overrides` for pnpm) so only one copy resolves.
High CVE: [GHSA redacted] desktop/windows/pnpm-lock.yaml— ip-address 10.2.0: [GHSA redacted] — ip-address is not declared in this repo's manifests: it is pulled in transitively, so upgrade the dependency that requires it (or pin ip-address to 10.3.1 with an `overrides` entry (`pnpm.overrides` for pnpm)). This is 1 of 3 advisories with a published fix this scan raises against ip-address 10.2.0, and their fixed versions do not agree — anything below 10.3.1 still leaves at least one of them open. Take this package to 10.3.1 or later: that is the floor for the package, not this row's target alone. (in 3 dependency files: desktop/windows/pnpm-lock.yaml, package-lock.json, sdks/react-native/package-lock.json) This one row stands for the 3 advisories this scan raises against ip-address 10.2.0: [GHSA redacted], [GHSA redacted], [GHSA redacted].
High CVE: [GHSA redacted] plugins/zapier/omi-zap/package-lock.json— js-yaml 3.14.2: [GHSA redacted] — js-yaml is not declared by the manifests that govern this lockfile: it is pulled in transitively here (the ~5.2.2 this repo declares for it belongs to a separate install root and does not reach this one), so upgrade the dependency that requires it (or pin js-yaml to 3.15.0 with an `overrides` entry (`pnpm.overrides` for pnpm)). This one row stands for the 2 advisories this scan raises against js-yaml 3.14.2: [GHSA redacted], [GHSA redacted].
High CVE: [GHSA redacted] backend/modal/requirements.txt— langsmith 0.8.5: [GHSA redacted] — langsmith is not declared by the manifests that govern this lockfile: it is pulled in transitively here (the 0.8.5 this repo declares for it belongs to a separate install root and does not reach this one), so upgrade the dependency that requires it (or raise its floor in your own manifest and re-resolve (`uv lock --upgrade-package langsmith`)). (in 4 dependency files)
High CVE: [GHSA redacted] backend/diarizer/requirements.txt— pillow 12.2.0: [GHSA redacted] — upgrade to 12.3.0. This one row stands for the 13 advisories this scan raises against pillow 12.2.0: [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted].
High CVE: [GHSA redacted] plugins/omi-ms365-app/requirements.txt— python-multipart 0.0.12: [GHSA redacted] — upgrade to 0.0.18. This is 1 of 8 advisories with a published fix this scan raises against python-multipart 0.0.12, and their fixed versions do not agree — anything below 0.0.31 still leaves at least one of them open. Take this package to 0.0.31 or later: that is the floor for the package, not this row's target alone. This one row stands for the 8 advisories this scan raises against python-multipart 0.0.12: [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], PYSEC-2026-3037, PYSEC-2026-3040, PYSEC-2026-3041.
High CVE: [GHSA redacted] backend/parakeet/requirements.txt— python-multipart 0.0.18: [GHSA redacted] — upgrade to 0.0.30. This is 1 of 7 advisories with a published fix this scan raises against python-multipart 0.0.18, and their fixed versions do not agree — anything below 0.0.31 still leaves at least one of them open. Take this package to 0.0.31 or later: that is the floor for the package, not this row's target alone. This one row stands for the 7 advisories this scan raises against python-multipart 0.0.18: [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], PYSEC-2026-3037, PYSEC-2026-3040, PYSEC-2026-3041.
High CVE: [GHSA redacted] backend/diarizer/requirements.txt— python-multipart 0.0.27: [GHSA redacted] — upgrade to 0.0.30. This is 1 of 4 advisories with a published fix this scan raises against python-multipart 0.0.27, and their fixed versions do not agree — anything below 0.0.31 still leaves at least one of them open. Take this package to 0.0.31 or later: that is the floor for the package, not this row's target alone. This one row stands for the 4 advisories this scan raises against python-multipart 0.0.27: [GHSA redacted], PYSEC-2026-3037, PYSEC-2026-3040, PYSEC-2026-3041.
High CVE: [GHSA redacted] plugins/composio/requirements.txt— python-multipart 0.0.29: [GHSA redacted] — upgrade to 0.0.30. This is 1 of 4 advisories with a published fix this scan raises against python-multipart 0.0.29, and their fixed versions do not agree — anything below 0.0.31 still leaves at least one of them open. Take this package to 0.0.31 or later: that is the floor for the package, not this row's target alone. (in 7 dependency files) This one row stands for the 4 advisories this scan raises against python-multipart 0.0.29: [GHSA redacted], PYSEC-2026-3037, PYSEC-2026-3040, PYSEC-2026-3041.
High CVE: [GHSA redacted] backend/diarizer/requirements.txt— pytorch-lightning 2.6.0: [GHSA redacted] — no fixed version has been published yet. Track the advisory, and remove or replace pytorch-lightning if the exposure is not acceptable until one lands.
High CVE: [GHSA redacted] sdks/react-native/package-lock.json— semver 7.3.8: [GHSA redacted] — semver is not declared in this repo's manifests: it is pulled in transitively, so upgrade the dependency that requires it (or pin semver to 7.5.2 with an `overrides` entry (`pnpm.overrides` for pnpm)).
High CVE: [GHSA redacted] sdks/react-native/package-lock.json— semver 7.5.1: [GHSA redacted] — semver is not declared in this repo's manifests: it is pulled in transitively, so upgrade the dependency that requires it (or pin semver to 7.5.2 with an `overrides` entry (`pnpm.overrides` for pnpm)).
High CVE: [GHSA redacted] plugins/requirements.txt— soupsieve 2.8.3: [GHSA redacted] — upgrade to 2.8.4. This one row stands for the 2 advisories this scan raises against soupsieve 2.8.3: [GHSA redacted], [GHSA redacted].
High CVE: [GHSA redacted] backend/diarizer/requirements.txt— starlette 0.49.1: [GHSA redacted] — upgrade to 1.3.1. This is 1 of 5 advisories with a published fix this scan raises against starlette 0.49.1, and their fixed versions do not agree — anything below 1.3.1 still leaves at least one of them open. Take this package to 1.3.1 or later: that is the floor for the package, not this row's target alone. (in 2 dependency files: backend/diarizer/requirements.txt, plugins/requirements.txt) This one row stands for the 5 advisories this scan raises against starlette 0.49.1: [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], PYSEC-2026-248.
High CVE: [GHSA redacted] backend/modal/requirements.txt— starlette 0.49.3: [GHSA redacted] — starlette is not declared by the manifests that govern this lockfile: it is pulled in transitively here (the 0.49.1 this repo declares for it belongs to a separate install root and does not reach this one), so upgrade the dependency that requires it (or raise its floor in your own manifest and re-resolve (`uv lock --upgrade-package starlette`)). This is 1 of 5 advisories with a published fix this scan raises against starlette 0.49.3, and their fixed versions do not agree — anything below 1.3.1 still leaves at least one of them open. Take this package to 1.3.1 or later: that is the floor for the package, not this row's target alone. (in 4 dependency files) This one row stands for the 5 advisories this scan raises against starlette 0.49.3: [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted], PYSEC-2026-248.
High IaC: DS-0029 backend/Dockerfile— 'apt-get' missing '--no-install-recommends'
High IaC: DS-0002 backend/Dockerfile.desktop_backend— Image user should not be 'root' A container that starts as root runs your process with root's capabilities inside the namespace, so a compromise of the process starts from there. The step: create an unprivileged account in the image (`RUN adduser --system --no-create-home app`), give it ownership of the paths the process writes at runtime (`COPY --chown=` on those layers, or a `RUN chown -R`), and end the final stage with `USER app` so it is the default at start. Build stages that only compile can stay root; it is the stage that RUNS that needs the account. If the process genuinely requires root — it manages the container runtime, ptraces another process or opens raw devices — say so here rather than making a change that breaks it.
High IaC: DS-0002 backend/agent-proxy/Dockerfile— Image user should not be 'root' A container that starts as root runs your process with root's capabilities inside the namespace, so a compromise of the process starts from there. The step: create an unprivileged account in the image (`RUN useradd -r -M app` — or whatever this base image's account tooling is, `adduser` and `useradd` are not both present everywhere`), give it ownership of the paths the process writes at runtime (`COPY --chown=` on those layers, or a `RUN chown -R`), and end the final stage with `USER app` so it is the default at start. Build stages that only compile can stay root; it is the stage that RUNS that needs the account. If the process genuinely requires root — it manages the container runtime, ptraces another process or opens raw devices — say so here rather than making a change that breaks it.
High IaC: DS-0002 backend/agent_vm/Dockerfile— Image user should not be 'root' A container that starts as root runs your process with root's capabilities inside the namespace, so a compromise of the process starts from there. The step: create an unprivileged account in the image (`RUN useradd -r -M app` — or whatever this base image's account tooling is, `adduser` and `useradd` are not both present everywhere`), give it ownership of the paths the process writes at runtime (`COPY --chown=` on those layers, or a `RUN chown -R`), and end the final stage with `USER app` so it is the default at start. Build stages that only compile can stay root; it is the stage that RUNS that needs the account. If the process genuinely requires root — it manages the container runtime, ptraces another process or opens raw devices — say so here rather than making a change that breaks it.
High IaC: KSV-0014 backend/charts/agent-vm-reaper/prod_agent_vm_reaper_cronjob.yaml— Root file system is not read-only
High IaC: KSV-0014 backend/charts/backend-listen/dev_omi_backend_listen_cronjob.yaml— Root file system is not read-only
High IaC: KSV-0014 backend/charts/backend-listen/prod_omi_backend_listen_cronjob.yaml— Root file system is not read-only
High IaC: DS-0002 backend/diarizer/Dockerfile— Image user should not be 'root' A container that starts as root runs your process with root's capabilities inside the namespace, so a compromise of the process starts from there. The step: create an unprivileged account in the image (`RUN adduser --system --no-create-home app`), give it ownership of the paths the process writes at runtime (`COPY --chown=` on those layers, or a `RUN chown -R`), and end the final stage with `USER app` so it is the default at start. Build stages that only compile can stay root; it is the stage that RUNS that needs the account. If the process genuinely requires root — it manages the container runtime, ptraces another process or opens raw devices — say so here rather than making a change that breaks it.
High IaC: DS-0029 backend/diarizer/Dockerfile— 'apt-get' missing '--no-install-recommends'
High IaC: DS-0002 backend/modal/Dockerfile— Image user should not be 'root' A container that starts as root runs your process with root's capabilities inside the namespace, so a compromise of the process starts from there. The step: create an unprivileged account in the image (`RUN adduser --system --no-create-home app`), give it ownership of the paths the process writes at runtime (`COPY --chown=` on those layers, or a `RUN chown -R`), and end the final stage with `USER app` so it is the default at start. Build stages that only compile can stay root; it is the stage that RUNS that needs the account. If the process genuinely requires root — it manages the container runtime, ptraces another process or opens raw devices — say so here rather than making a change that breaks it.
High IaC: DS-0029 backend/modal/Dockerfile— 'apt-get' missing '--no-install-recommends'
High IaC: DS-0002 backend/modal/Dockerfile.memory_maintenance_job— Image user should not be 'root' A container that starts as root runs your process with root's capabilities inside the namespace, so a compromise of the process starts from there. The step: create an unprivileged account in the image (`RUN adduser --system --no-create-home app`), give it ownership of the paths the process writes at runtime (`COPY --chown=` on those layers, or a `RUN chown -R`), and end the final stage with `USER app` so it is the default at start. Build stages that only compile can stay root; it is the stage that RUNS that needs the account. If the process genuinely requires root — it manages the container runtime, ptraces another process or opens raw devices — say so here rather than making a change that breaks it.
High IaC: DS-0002 backend/modal/Dockerfile.notifications_job— Image user should not be 'root' A container that starts as root runs your process with root's capabilities inside the namespace, so a compromise of the process starts from there. The step: create an unprivileged account in the image (`RUN adduser --system --no-create-home app`), give it ownership of the paths the process writes at runtime (`COPY --chown=` on those layers, or a `RUN chown -R`), and end the final stage with `USER app` so it is the default at start. Build stages that only compile can stay root; it is the stage that RUNS that needs the account. If the process genuinely requires root — it manages the container runtime, ptraces another process or opens raw devices — say so here rather than making a change that breaks it.
High IaC: DS-0029 backend/modal/Dockerfile.notifications_job— 'apt-get' missing '--no-install-recommends'
High IaC: DS-0002 backend/nllb_translation/Dockerfile— Image user should not be 'root' A container that starts as root runs your process with root's capabilities inside the namespace, so a compromise of the process starts from there. The step: create an unprivileged account in the image (`RUN adduser --system --no-create-home app`), give it ownership of the paths the process writes at runtime (`COPY --chown=` on those layers, or a `RUN chown -R`), and end the final stage with `USER app` so it is the default at start. Build stages that only compile can stay root; it is the stage that RUNS that needs the account. If the process genuinely requires root — it manages the container runtime, ptraces another process or opens raw devices — say so here rather than making a change that breaks it.
High IaC: DS-0002 backend/parakeet/Dockerfile— Image user should not be 'root' A container that starts as root runs your process with root's capabilities inside the namespace, so a compromise of the process starts from there. The step: create an unprivileged account in the image (`RUN useradd -r -M app` — or whatever this base image's account tooling is, `adduser` and `useradd` are not both present everywhere`), give it ownership of the paths the process writes at runtime (`COPY --chown=` on those layers, or a `RUN chown -R`), and end the final stage with `USER app` so it is the default at start. Build stages that only compile can stay root; it is the stage that RUNS that needs the account. If the process genuinely requires root — it manages the container runtime, ptraces another process or opens raw devices — say so here rather than making a change that breaks it.
High IaC: DS-0002 backend/parakeet/Dockerfile.nim— Image user should not be 'root' A container that starts as root runs your process with root's capabilities inside the namespace, so a compromise of the process starts from there. The step: create an unprivileged account in the image (`RUN useradd -r -M app` — or whatever this base image's account tooling is, `adduser` and `useradd` are not both present everywhere`), give it ownership of the paths the process writes at runtime (`COPY --chown=` on those layers, or a `RUN chown -R`), and end the final stage with `USER app` so it is the default at start. Build stages that only compile can stay root; it is the stage that RUNS that needs the account. If the process genuinely requires root — it manages the container runtime, ptraces another process or opens raw devices — say so here rather than making a change that breaks it.
High IaC: DS-0029 backend/pusher/Dockerfile— 'apt-get' missing '--no-install-recommends'
High IaC: DS-0002 backend/testing/chaos-oom/Dockerfile.harness— Image user should not be 'root' A container that starts as root runs your process with root's capabilities inside the namespace, so a compromise of the process starts from there. The step: create an unprivileged account in the image (`RUN useradd -r -M app` — or whatever this base image's account tooling is, `adduser` and `useradd` are not both present everywhere`), give it ownership of the paths the process writes at runtime (`COPY --chown=` on those layers, or a `RUN chown -R`), and end the final stage with `USER app` so it is the default at start. Build stages that only compile can stay root; it is the stage that RUNS that needs the account. If the process genuinely requires root — it manages the container runtime, ptraces another process or opens raw devices — say so here rather than making a change that breaks it.
High IaC: DS-0002 mcp/Dockerfile— Image user should not be 'root' A container that starts as root runs your process with root's capabilities inside the namespace, so a compromise of the process starts from there. The step: create an unprivileged account in the image (`RUN adduser --system --no-create-home app`), give it ownership of the paths the process writes at runtime (`COPY --chown=` on those layers, or a `RUN chown -R`), and end the final stage with `USER app` so it is the default at start. Build stages that only compile can stay root; it is the stage that RUNS that needs the account. If the process genuinely requires root — it manages the container runtime, ptraces another process or opens raw devices — say so here rather than making a change that breaks it.
High IaC: DS-0029 mcp/Dockerfile— 'apt-get' missing '--no-install-recommends'
High IaC: DS-0002 plugins/Dockerfile— Image user should not be 'root' A container that starts as root runs your process with root's capabilities inside the namespace, so a compromise of the process starts from there. The step: create an unprivileged account in the image (`RUN adduser --system --no-create-home app`), give it ownership of the paths the process writes at runtime (`COPY --chown=` on those layers, or a `RUN chown -R`), and end the final stage with `USER app` so it is the default at start. Build stages that only compile can stay root; it is the stage that RUNS that needs the account. If the process genuinely requires root — it manages the container runtime, ptraces another process or opens raw devices — say so here rather than making a change that breaks it.
High IaC: DS-0029 plugins/Dockerfile— 'apt-get' missing '--no-install-recommends'
High IaC: DS-0002 plugins/Dockerfile.datadog— Image user should not be 'root' A container that starts as root runs your process with root's capabilities inside the namespace, so a compromise of the process starts from there. The step: create an unprivileged account in the image (`RUN adduser --system --no-create-home app`), give it ownership of the paths the process writes at runtime (`COPY --chown=` on those layers, or a `RUN chown -R`), and end the final stage with `USER app` so it is the default at start. Build stages that only compile can stay root; it is the stage that RUNS that needs the account. If the process genuinely requires root — it manages the container runtime, ptraces another process or opens raw devices — say so here rather than making a change that breaks it.
High IaC: DS-0029 plugins/Dockerfile.datadog— 'apt-get' missing '--no-install-recommends'
Leaked secret: gcp-api-key config/public-build-values.json:6— gcp-api-key detected. Treat the value as compromised: it is readable by everyone who has ever had the repository, and deleting the line does not un-publish it. In order — (1) REVOKE it at whatever issued it and issue a replacement, which is the only step that actually closes the exposure; (2) load the replacement at run time from your platform's secret store or the process environment instead of from the tree, so no future value is committable; (3) remove the file or line and add its path to the repository's ignore rules, so it cannot come back; (4) if the value was ever live, purge it from the history as well, since a clone taken before the deletion still carries it. If this is instead a FIXTURE — key material generated for tests and valid nowhere — then the exposure is nil and the fix is to make that legible: generate it in test setup, or keep it under a test-data path, so a reader (and this scan) can tell it from the real thing.
Leaked secret: gcp-api-key desktop/macos/run.sh:118— gcp-api-key detected. Treat the value as compromised: it is readable by everyone who has ever had the repository, and deleting the line does not un-publish it. In order — (1) REVOKE it at whatever issued it and issue a replacement, which is the only step that actually closes the exposure; (2) load the replacement at run time from your platform's secret store or the process environment instead of from the tree, so no future value is committable; (3) remove the file or line and add its path to the repository's ignore rules, so it cannot come back; (4) if the value was ever live, purge it from the history as well, since a clone taken before the deletion still carries it. If this is instead a FIXTURE — key material generated for tests and valid nowhere — then the exposure is nil and the fix is to make that legible: generate it in test setup, or keep it under a test-data path, so a reader (and this scan) can tell it from the real thing.
Leaked secret: high-entropy-secret desktop/macos/run.sh:118— high-entropy-secret detected. Treat the value as compromised: it is readable by everyone who has ever had the repository, and deleting the line does not un-publish it. In order — (1) REVOKE it at whatever issued it and issue a replacement, which is the only step that actually closes the exposure; (2) load the replacement at run time from your platform's secret store or the process environment instead of from the tree, so no future value is committable; (3) remove the file or line and add its path to the repository's ignore rules, so it cannot come back; (4) if the value was ever live, purge it from the history as well, since a clone taken before the deletion still carries it. If this is instead a FIXTURE — key material generated for tests and valid nowhere — then the exposure is nil and the fix is to make that legible: generate it in test setup, or keep it under a test-data path, so a reader (and this scan) can tell it from the real thing.
Leaked secret: high-entropy-secret backend/routers/mcp_sse.py:100— high-entropy-secret detected. Treat the value as compromised: it is readable by everyone who has ever had the repository, and deleting the line does not un-publish it. In order — (1) REVOKE it at whatever issued it and issue a replacement, which is the only step that actually closes the exposure; (2) load the replacement at run time from your platform's secret store or the process environment instead of from the tree, so no future value is committable; (3) remove the file or line and add its path to the repository's ignore rules, so it cannot come back; (4) if the value was ever live, purge it from the history as well, since a clone taken before the deletion still carries it. If this is instead a FIXTURE — key material generated for tests and valid nowhere — then the exposure is nil and the fix is to make that legible: generate it in test setup, or keep it under a test-data path, so a reader (and this scan) can tell it from the real thing.
Leaked secret: gcp-api-key sdks/python-cli/omi_cli/auth/oauth.py:55— gcp-api-key detected. Treat the value as compromised: it is readable by everyone who has ever had the repository, and deleting the line does not un-publish it. In order — (1) REVOKE it at whatever issued it and issue a replacement, which is the only step that actually closes the exposure; (2) load the replacement at run time from your platform's secret store or the process environment instead of from the tree, so no future value is committable; (3) remove the file or line and add its path to the repository's ignore rules, so it cannot come back; (4) if the value was ever live, purge it from the history as well, since a clone taken before the deletion still carries it. If this is instead a FIXTURE — key material generated for tests and valid nowhere — then the exposure is nil and the fix is to make that legible: generate it in test setup, or keep it under a test-data path, so a reader (and this scan) can tell it from the real thing.
Leaked secret: high-entropy-secret sdks/python-cli/omi_cli/auth/oauth.py:55— high-entropy-secret detected. Treat the value as compromised: it is readable by everyone who has ever had the repository, and deleting the line does not un-publish it. In order — (1) REVOKE it at whatever issued it and issue a replacement, which is the only step that actually closes the exposure; (2) load the replacement at run time from your platform's secret store or the process environment instead of from the tree, so no future value is committable; (3) remove the file or line and add its path to the repository's ignore rules, so it cannot come back; (4) if the value was ever live, purge it from the history as well, since a clone taken before the deletion still carries it. If this is instead a FIXTURE — key material generated for tests and valid nowhere — then the exposure is nil and the fix is to make that legible: generate it in test setup, or keep it under a test-data path, so a reader (and this scan) can tell it from the real thing.
Leaked secret: private-key omi/firmware/bootloader/mcuboot/root-rsa-2048.pem:1— private-key detected. Treat the value as compromised: it is readable by everyone who has ever had the repository, and deleting the line does not un-publish it. In order — (1) REVOKE it at whatever issued it and issue a replacement, which is the only step that actually closes the exposure; (2) load the replacement at run time from your platform's secret store or the process environment instead of from the tree, so no future value is committable; (3) remove the file or line and add its path to the repository's ignore rules, so it cannot come back; (4) if the value was ever live, purge it from the history as well, since a clone taken before the deletion still carries it. If this is instead a FIXTURE — key material generated for tests and valid nowhere — then the exposure is nil and the fix is to make that legible: generate it in test setup, or keep it under a test-data path, so a reader (and this scan) can tell it from the real thing.
Leaked secret: private-key omi/firmware/bootloader/mcuboot/enc-rsa2048-priv.pem:1— private-key detected. Treat the value as compromised: it is readable by everyone who has ever had the repository, and deleting the line does not un-publish it. In order — (1) REVOKE it at whatever issued it and issue a replacement, which is the only step that actually closes the exposure; (2) load the replacement at run time from your platform's secret store or the process environment instead of from the tree, so no future value is committable; (3) remove the file or line and add its path to the repository's ignore rules, so it cannot come back; (4) if the value was ever live, purge it from the history as well, since a clone taken before the deletion still carries it. If this is instead a FIXTURE — key material generated for tests and valid nowhere — then the exposure is nil and the fix is to make that legible: generate it in test setup, or keep it under a test-data path, so a reader (and this scan) can tell it from the real thing.
D38 · OSV Dependency Vulnerabilities· High vulnerability · ×4
High vulnerability: [GHSA redacted] backend/agent-proxy/requirements.txt— cryptography 46.0.7: [GHSA redacted] — upgrade to 48.0.1. This is 1 of 4 advisories with a published fix this scan raises against cryptography 46.0.7, and their fixed versions do not agree — anything below 50.0.0 still leaves at least one of them open. Take this package to 50.0.0 or later: that is the floor for the package, not this row's target alone. This one row stands for the 4 advisories this scan raises against cryptography 46.0.7: [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted].
High vulnerability: [GHSA redacted] plugins/requirements.txt— cryptography 48.0.0: [GHSA redacted] — upgrade to 48.0.1. This one row stands for the 4 advisories this scan raises against cryptography 48.0.0: [GHSA redacted], [GHSA redacted], [GHSA redacted], [GHSA redacted].
High vulnerability: [GHSA redacted] backend/modal/requirements.txt— gitpython 3.1.55: [GHSA redacted] — gitpython is not declared by the manifests that govern this lockfile: it is pulled in transitively here (the 3.1.55 this repo declares for it belongs to a separate install root and does not reach this one), so upgrade the dependency that requires it (or raise its floor in your own manifest and re-resolve (`uv lock --upgrade-package gitpython`)). This is 1 of 3 advisories with a published fix this scan raises against gitpython 3.1.55, and their fixed versions do not agree — anything below 3.1.57 still leaves at least one of them open. Take this package to 3.1.57 or later: that is the floor for the package, not this row's target alone. (in 2 dependency files: backend/modal/requirements.txt, backend/requirements.txt) This one row stands for the 3 advisories this scan raises against gitpython 3.1.55: [GHSA redacted], [GHSA redacted], [GHSA redacted].
High vulnerability: [GHSA redacted] desktop/windows/pnpm-lock.yaml— react-router 7.18.2: [GHSA redacted] — react-router is not declared in this repo's manifests: it is pulled in transitively, so upgrade the dependency that requires it (— 8.3.0 is a MAJOR ahead of the resolved 7.18.2, so an `overrides` pin would force a breaking version under a dependent written against 7.18.2; upgrading the declaring package is the remedy, and where no patched release exists in your major line, record the exposure instead).
Boundary-crossing change coupling: message.dart ↔ inventory_app_client_schemas.py app/lib/backend/schema/message.dart— `app/lib/backend/schema/message.dart` (context schema) and `backend/scripts/inventory_app_client_schemas.py` (context scripts) sit in DIFFERENT parts of the tree yet change together 73% of the time (8 of the 11 commits that touched whichever of the two files changed less often, counting a file under its earlier names as well) — the bounded-context boundary may be in the wrong place, or one context is leaking into the other. This is the behavioural boundary violation a static scan can't see.
Boundary-crossing change coupling: conversation_finalization.py ↔ lifecycle.py backend/services/conversation_finalization.py— `backend/services/conversation_finalization.py` (context services) and `backend/utils/conversations/lifecycle.py` (context utils) sit in DIFFERENT parts of the tree yet change together 54% of the time (7 of the 13 commits that touched whichever of the two files changed less often, counting a file under its earlier names as well) — the bounded-context boundary may be in the wrong place, or one context is leaking into the other. This is the behavioural boundary violation a static scan can't see.
High: watchdog-sensitive-personal-data-in-log desktop/windows/src/main/glow/glowWindow.ts:186— Special-category or directly-identifying personal data appears to be written to a log or console sink. Logs are copied, shipped to third parties and retained far longer than the data itself is lawfully needed. Log a stable pseudonymous reference instead of the value.
Duplicated block (7 lines × 2) .github/scripts/check_deployment_secret_boundary.py:317— .github/scripts/check_deployment_secret_boundary.py:317-323 | .github/scripts/check_public_build_contract.py:650-656 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (7 lines × 2) app/bee_audio_capture.py:186— app/bee_audio_capture.py:186-192 | app/bee_audio_capture.py:274-280 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (7 lines × 2) backend/agent-proxy/main.py:220— backend/agent-proxy/main.py:220-226 | backend/routers/agent_tools.py:99-105 — the copies span different directories, so extracting a shared function means choosing where it lives: put it somewhere both call sites can already reach — a location they all depend on today, or a new shared one if there is none — and call it from each site; until then, every change has to be made twice.
Duplicated block (7 lines × 2) backend/agent-proxy/main.py:350— backend/agent-proxy/main.py:350-356 | backend/agent-proxy/main.py:393-399 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (7 lines × 2) backend/database/candidates.py:654— backend/database/candidates.py:654-660 | backend/database/candidates.py:907-913 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/database/candidates.py:654` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (7 lines × 2) backend/database/chat.py:590— backend/database/chat.py:590-596 | backend/database/chat.py:620-626 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (7 lines × 2) backend/database/chat_first_intents.py:323— backend/database/chat_first_intents.py:323-329 | backend/database/chat_first_intents.py:402-408 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (7 lines × 2) backend/database/chat_first_intents.py:477— backend/database/chat_first_intents.py:477-483 | backend/database/chat_first_intents.py:575-581 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (7 lines × 2) backend/database/conversation_finalization_jobs.py:524— backend/database/conversation_finalization_jobs.py:524-530 | backend/database/conversation_finalization_jobs.py:735-741 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (7 lines × 2) backend/database/conversation_finalization_jobs.py:1214— backend/database/conversation_finalization_jobs.py:1214-1220 | backend/database/conversation_finalization_jobs.py:1259-1265 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (7 lines × 2) backend/database/conversations.py:646— backend/database/conversations.py:646-652 | backend/database/conversations.py:693-702 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/database/conversations.py:646` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (7 lines × 2) backend/database/dev_api_key.py:232— backend/database/dev_api_key.py:232-238 | backend/database/mcp_api_key.py:329-335 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (7 lines × 2) backend/database/llm_usage.py:212— backend/database/llm_usage.py:212-218 | backend/database/llm_usage.py:280-286 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (7 lines × 2) backend/database/memories.py:212— backend/database/memories.py:212-219 | backend/database/memories.py:240-246 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (7 lines × 2) backend/database/memory_outbox_worker.py:595— backend/database/memory_outbox_worker.py:595-601 | backend/database/memory_outbox_worker.py:607-613 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/database/memory_outbox_worker.py:595` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (7 lines × 2) backend/database/memory_vector_repair_outbox_telemetry.py:85— backend/database/memory_vector_repair_outbox_telemetry.py:85-91 | backend/database/memory_vector_repair_outbox_telemetry.py:192-198 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (7 lines × 2) backend/database/memory_vector_repair_outbox_worker.py:363— backend/database/memory_vector_repair_outbox_worker.py:363-369 | backend/database/memory_vector_repair_outbox_worker.py:384-390 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (7 lines × 2) backend/database/workstreams.py:465— backend/database/workstreams.py:465-471 | backend/database/workstreams.py:863-869 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (7 lines × 2) backend/database/workstreams.py:996— backend/database/workstreams.py:996-1002 | backend/database/workstreams.py:1198-1204 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (7 lines × 2) backend/llm_gateway/gateway/providers.py:559— backend/llm_gateway/gateway/providers.py:559-565 | backend/utils/llm/desktop_llm_stub.py:53-59 — the copies span different directories, so extracting a shared function means choosing where it lives: put it somewhere both call sites can already reach — a location they all depend on today, or a new shared one if there is none — and call it from each site; until then, every change has to be made twice. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (7 lines × 2) backend/migrations/001_enhanced_protection_default.py:75— backend/migrations/001_enhanced_protection_default.py:75-81 | backend/migrations/001_enhanced_protection_default.py:83-89 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (7 lines × 2) backend/routers/apps.py:2125— backend/routers/apps.py:2125-2131 | backend/routers/oauth.py:253-259 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once.
Duplicated block (7 lines × 2) backend/routers/desktop_agent_vm.py:428— backend/routers/desktop_agent_vm.py:428-434 | backend/routers/desktop_agent_vm.py:439-445 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (7 lines × 2) backend/routers/desktop_chat.py:355— backend/routers/desktop_chat.py:355-361 | backend/routers/desktop_chat.py:376-382 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (7 lines × 2) backend/routers/developer.py:341— backend/routers/developer.py:341-350 | backend/routers/mcp.py:351-357 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once.
Duplicated block (10 lines × 2) .github/scripts/check_opentofu_development_wif_pilot.py:250— .github/scripts/check_opentofu_development_wif_pilot.py:250-259 | .github/scripts/check_opentofu_development_wif_pilot.py:335-344 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (10 lines × 2) .github/scripts/check_public_build_contract.py:157— .github/scripts/check_public_build_contract.py:157-166 | .github/scripts/check_public_build_contract.py:195-204 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `.github/scripts/check_public_build_contract.py:157` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (10 lines × 2) backend/agent-proxy/main.py:237— backend/agent-proxy/main.py:237-246 | backend/routers/agent_tools.py:114-123 — the copies span different directories, so extracting a shared function means choosing where it lives: put it somewhere both call sites can already reach — a location they all depend on today, or a new shared one if there is none — and call it from each site; until then, every change has to be made twice.
Duplicated block (10 lines × 2) backend/agent-proxy/main.py:270— backend/agent-proxy/main.py:270-279 | backend/routers/agent_tools.py:147-156 — the copies span different directories, so extracting a shared function means choosing where it lives: put it somewhere both call sites can already reach — a location they all depend on today, or a new shared one if there is none — and call it from each site; until then, every change has to be made twice. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (10 lines × 2) backend/database/action_items.py:225— backend/database/action_items.py:225-234 | backend/database/action_items.py:330-339 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (10 lines × 2) backend/database/announcements.py:149— backend/database/announcements.py:149-159 | backend/database/announcements.py:421-430 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/database/announcements.py:149` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (10 lines × 2) backend/database/chat.py:186— backend/database/chat.py:186-195 | backend/database/chat.py:245-254 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (10 lines × 2) backend/database/conversations.py:363— backend/database/conversations.py:363-372 | backend/database/conversations.py:433-442 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (10 lines × 2) backend/database/conversations.py:706— backend/database/conversations.py:706-715 | backend/database/conversations.py:1339-1348 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (10 lines × 2) backend/database/helpers.py:235— backend/database/helpers.py:235-244 | backend/database/helpers.py:298-308 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/database/helpers.py:235` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (10 lines × 2) backend/database/memory_vector_metadata.py:134— backend/database/memory_vector_metadata.py:134-143 | backend/database/memory_vector_metadata.py:163-172 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/database/memory_vector_metadata.py:134` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (10 lines × 2) backend/database/sync_jobs.py:307— backend/database/sync_jobs.py:307-316 | backend/database/sync_jobs.py:410-420 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (10 lines × 2) backend/dependencies.py:124— backend/dependencies.py:124-133 | backend/dependencies.py:154-163 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/dependencies.py:124` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (10 lines × 2) backend/migrations/001_enhanced_protection_default.py:152— backend/migrations/001_enhanced_protection_default.py:152-161 | backend/migrations/002_populate_historical_usage.py:265-274 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. Read the line range as the matched WINDOW rather than a finished unit: at `backend/migrations/001_enhanced_protection_default.py:152` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (10 lines × 2) backend/migrations/002_populate_historical_usage.py:239— backend/migrations/002_populate_historical_usage.py:239-248 | backend/migrations/003_report_top_transcription_users.py:111-120 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once.
Duplicated block (10 lines × 2) backend/models/memories.py:439— backend/models/memories.py:439-448 | backend/models/memories.py:482-491 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/models/memories.py:439` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (10 lines × 2) backend/routers/chat_first.py:275— backend/routers/chat_first.py:275-284 | backend/routers/chat_first.py:297-307 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/routers/chat_first.py:275` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (10 lines × 2) backend/routers/mcp_sse.py:882— backend/routers/mcp_sse.py:882-891 | backend/routers/mcp_sse.py:1111-1120 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/routers/mcp_sse.py:882` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (10 lines × 2) backend/routers/mcp_sse.py:912— backend/routers/mcp_sse.py:912-921 | backend/routers/mcp_sse.py:964-973 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/routers/mcp_sse.py:912` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (10 lines × 2) backend/routers/mcp_sse.py:1081— backend/routers/mcp_sse.py:1081-1090 | backend/routers/mcp_sse.py:1187-1197 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/routers/mcp_sse.py:1187` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (10 lines × 2) backend/routers/mcp_sse.py:1286— backend/routers/mcp_sse.py:1286-1295 | backend/routers/mcp_sse.py:1305-1314 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (10 lines × 2) backend/routers/payment.py:426— backend/routers/payment.py:426-435 | backend/routers/payment.py:448-457 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/routers/payment.py:426` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (10 lines × 2) backend/routers/task_integrations.py:763— backend/routers/task_integrations.py:763-772 | backend/routers/task_integrations.py:816-825 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (10 lines × 2) backend/routers/users.py:1264— backend/routers/users.py:1264-1273 | backend/routers/users.py:1286-1295 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/routers/users.py:1264` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (10 lines × 2) backend/scripts/export_openapi.py:438— backend/scripts/export_openapi.py:438-447 | backend/scripts/route_policy_inventory.py:755-764 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once.
Duplicated block (6 lines × 2) .github/scripts/check-codemagic-tag-intake.py:461— .github/scripts/check-codemagic-tag-intake.py:461-466 | backend/scripts/agent_vm_release.py:84-89 — the copies span different directories, so extracting a shared function means choosing where it lives: put it somewhere both call sites can already reach — a location they all depend on today, or a new shared one if there is none — and call it from each site; until then, every change has to be made twice.
Duplicated block (6 lines × 2) .github/scripts/verify_release_eligibility.py:55— .github/scripts/verify_release_eligibility.py:55-60 | backend/scripts/resolve_cloud_run_tagged_url.py:129-134 — the copies span different directories, so extracting a shared function means choosing where it lives: put it somewhere both call sites can already reach — a location they all depend on today, or a new shared one if there is none — and call it from each site; until then, every change has to be made twice. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (6 lines × 2) backend/agent-proxy/main.py:205— backend/agent-proxy/main.py:205-210 | backend/routers/agent_tools.py:82-87 — the copies span different directories, so extracting a shared function means choosing where it lives: put it somewhere both call sites can already reach — a location they all depend on today, or a new shared one if there is none — and call it from each site; until then, every change has to be made twice. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (6 lines × 2) backend/agent-proxy/main.py:230— backend/agent-proxy/main.py:230-235 | backend/routers/agent_tools.py:107-112 — the copies span different directories, so extracting a shared function means choosing where it lives: put it somewhere both call sites can already reach — a location they all depend on today, or a new shared one if there is none — and call it from each site; until then, every change has to be made twice.
Duplicated block (6 lines × 2) backend/agent-proxy/main.py:397— backend/agent-proxy/main.py:397-402 | backend/routers/desktop_agent_vm.py:257-262 — the copies span different directories, so extracting a shared function means choosing where it lives: put it somewhere both call sites can already reach — a location they all depend on today, or a new shared one if there is none — and call it from each site; until then, every change has to be made twice. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (6 lines × 2) backend/database/action_items.py:41— backend/database/action_items.py:41-46 | backend/database/action_items.py:53-58 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (6 lines × 2) backend/database/advice.py:59— backend/database/advice.py:59-64 | backend/database/screen_activity.py:79-84 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once.
Duplicated block (6 lines × 2) backend/database/announcements.py:129— backend/database/announcements.py:129-136 | backend/database/announcements.py:196-201 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/database/announcements.py:129` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (6 lines × 2) backend/database/apps.py:463— backend/database/apps.py:463-468 | backend/database/apps.py:474-479 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (6 lines × 2) backend/database/helpers.py:71— backend/database/helpers.py:71-76 | backend/database/helpers.py:105-110 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/database/helpers.py:71` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (6 lines × 2) backend/database/memories.py:277— backend/database/memories.py:277-282 | backend/database/memories.py:613-618 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (6 lines × 2) backend/database/memory_apply_store.py:1161— backend/database/memory_apply_store.py:1161-1166 | backend/database/memory_apply_store.py:1209-1214 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (6 lines × 2) backend/database/memory_ledger.py:313— backend/database/memory_ledger.py:313-318 | backend/database/memory_ledger.py:366-371 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (6 lines × 2) backend/llm_gateway/gateway/providers.py:699— backend/llm_gateway/gateway/providers.py:699-704 | backend/llm_gateway/gateway/providers.py:753-758 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (6 lines × 2) backend/parakeet/stream_handler.py:771— backend/parakeet/stream_handler.py:771-776 | backend/parakeet/transcribe.py:357-362 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once.
Duplicated block (6 lines × 2) backend/routers/desktop_agent_vm.py:635— backend/routers/desktop_agent_vm.py:635-640 | backend/routers/desktop_agent_vm.py:650-655 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (6 lines × 2) backend/routers/users.py:1553— backend/routers/users.py:1553-1559 | backend/routers/users.py:1719-1724 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (6 lines × 2) backend/scripts/benchmark_nllb_performance.py:298— backend/scripts/benchmark_nllb_performance.py:298-303 | backend/scripts/benchmark_translation.py:484-489 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/benchmark_nllb_performance.py:298` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (6 lines × 2) backend/scripts/cutover_evidence_readiness.py:210— backend/scripts/cutover_evidence_readiness.py:210-215 | backend/scripts/rollout_schema_readiness.py:214-219 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (6 lines × 2) backend/scripts/export_openapi.py:936— backend/scripts/export_openapi.py:936-941 | backend/scripts/export_openapi.py:971-976 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (6 lines × 2) backend/scripts/pinecone_repair_validation_readiness.py:99— backend/scripts/pinecone_repair_validation_readiness.py:99-104 | backend/scripts/shared_ns2_legacy_isolation_readiness.py:99-104 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/pinecone_repair_validation_readiness.py:99` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (6 lines × 2) backend/scripts/runtime_env_durable_dispatch_contracts.py:97— backend/scripts/runtime_env_durable_dispatch_contracts.py:97-102 | backend/scripts/runtime_env_durable_dispatch_contracts.py:124-129 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (6 lines × 2) backend/scripts/stt/m_benchmark_streaming.py:413— backend/scripts/stt/m_benchmark_streaming.py:413-418 | backend/scripts/stt/m_benchmark_streaming.py:422-428 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/stt/m_benchmark_streaming.py:413` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (6 lines × 2) backend/scripts/stt/n_benchmark_02_prerecorded.py:330— backend/scripts/stt/n_benchmark_02_prerecorded.py:330-335 | backend/scripts/stt/o_benchmark_02_streaming.py:349-354 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (6 lines × 2) backend/scripts/stt/o_benchmark_02_streaming.py:319— backend/scripts/stt/o_benchmark_02_streaming.py:319-324 | backend/scripts/stt/o_benchmark_02_streaming.py:330-336 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/stt/o_benchmark_02_streaming.py:319` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (9 lines × 2) .github/scripts/desktop-release-source-identity.py:61— .github/scripts/desktop-release-source-identity.py:61-69 | .github/scripts/resolve-desktop-changelog-sync.py:44-52 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (9 lines × 2) backend/agent-proxy/main.py:252— backend/agent-proxy/main.py:252-260 | backend/routers/agent_tools.py:129-137 — the copies span different directories, so extracting a shared function means choosing where it lives: put it somewhere both call sites can already reach — a location they all depend on today, or a new shared one if there is none — and call it from each site; until then, every change has to be made twice. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (9 lines × 2) backend/database/chat.py:309— backend/database/chat.py:309-320 | backend/database/chat.py:1209-1217 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (9 lines × 2) backend/database/chat.py:1123— backend/database/chat.py:1123-1131 | backend/database/chat.py:1174-1182 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (9 lines × 2) backend/database/conversations.py:632— backend/database/conversations.py:632-642 | backend/database/conversations.py:681-689 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (9 lines × 2) backend/database/dev_api_key.py:129— backend/database/dev_api_key.py:129-137 | backend/database/mcp_api_key.py:221-229 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (9 lines × 2) backend/database/goals.py:537— backend/database/goals.py:537-545 | backend/database/goals.py:624-632 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/database/goals.py:537` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (9 lines × 2) backend/database/memory_apply_store.py:1094— backend/database/memory_apply_store.py:1094-1102 | backend/database/memory_apply_store.py:1113-1121 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/database/memory_apply_store.py:1094` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (9 lines × 2) backend/database/memory_apply_store.py:1183— backend/database/memory_apply_store.py:1183-1191 | backend/database/memory_apply_store.py:1234-1244 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (9 lines × 2) backend/database/task_recommendations.py:794— backend/database/task_recommendations.py:794-802 | backend/database/task_recommendations.py:901-909 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/database/task_recommendations.py:794` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (9 lines × 2) backend/database/vector_db.py:722— backend/database/vector_db.py:722-731 | backend/database/vector_db.py:1124-1132 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (9 lines × 2) backend/parakeet/gpu_worker.py:231— backend/parakeet/gpu_worker.py:231-239 | backend/parakeet/gpu_worker.py:249-257 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (9 lines × 2) backend/parakeet/stream_handler.py:759— backend/parakeet/stream_handler.py:759-767 | backend/parakeet/transcribe.py:344-352 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (9 lines × 2) backend/routers/chat.py:1558— backend/routers/chat.py:1558-1566 | backend/routers/users.py:801-810 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once.
Duplicated block (9 lines × 2) backend/routers/pusher.py:383— backend/routers/pusher.py:383-391 | backend/testing/chaos-oom/pusher_fixed.py:346-354 — the copies span different directories, so extracting a shared function means choosing where it lives: put it somewhere both call sites can already reach — a location they all depend on today, or a new shared one if there is none — and call it from each site; until then, every change has to be made twice.
Duplicated block (9 lines × 2) backend/routers/task_integrations.py:776— backend/routers/task_integrations.py:776-784 | backend/routers/task_integrations.py:829-837 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (9 lines × 2) backend/scripts/backfill_developer_api_key_memory_grants.py:101— backend/scripts/backfill_developer_api_key_memory_grants.py:101-109 | backend/scripts/backfill_mcp_key_full_access.py:85-93 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once.
Duplicated block (9 lines × 2) backend/scripts/check_app_client_openapi_compatibility.py:195— backend/scripts/check_app_client_openapi_compatibility.py:195-203 | backend/scripts/check_app_client_openapi_compatibility.py:227-235 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (9 lines × 2) backend/scripts/check_module_stub_pollution.py:237— backend/scripts/check_module_stub_pollution.py:237-246 | backend/scripts/scan_import_time_side_effects.py:412-420 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once.
Duplicated block (9 lines × 2) backend/scripts/export_openapi.py:415— backend/scripts/export_openapi.py:415-423 | backend/testing/hermetic_network.py:16-25 — the copies span different directories, so extracting a shared function means choosing where it lives: put it somewhere both call sites can already reach — a location they all depend on today, or a new shared one if there is none — and call it from each site; until then, every change has to be made twice. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (9 lines × 2) backend/scripts/generate_swift_openapi_types.py:318— backend/scripts/generate_swift_openapi_types.py:318-328 | backend/scripts/generate_swift_openapi_types.py:476-484 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/generate_swift_openapi_types.py:318` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (9 lines × 2) backend/scripts/generate_swift_openapi_types.py:609— backend/scripts/generate_swift_openapi_types.py:609-621 | backend/scripts/generate_ts_openapi_types.py:198-206 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (9 lines × 2) backend/scripts/pinecone_repair_validation_readiness.py:84— backend/scripts/pinecone_repair_validation_readiness.py:84-93 | backend/scripts/pinecone_repair_validation_readiness.py:94-102 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/pinecone_repair_validation_readiness.py:84` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (9 lines × 2) backend/scripts/stt/k_compare_transcripts_performance.py:212— backend/scripts/stt/k_compare_transcripts_performance.py:212-220 | backend/scripts/stt/k_compare_transcripts_performance.py:236-246 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/stt/k_compare_transcripts_performance.py:212` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (9 lines × 2) backend/scripts/stt/l_benchmark_prerecorded.py:126— backend/scripts/stt/l_benchmark_prerecorded.py:126-134 | backend/scripts/stt/m_benchmark_streaming.py:131-139 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/stt/l_benchmark_prerecorded.py:126` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (11 lines × 2) .github/scripts/check-release-process-guards.py:813— .github/scripts/check-release-process-guards.py:813-823 | .github/scripts/check-release-process-guards.py:832-842 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (11 lines × 2) .github/scripts/check_analytics_reachability.py:135— .github/scripts/check_analytics_reachability.py:135-145 | .github/scripts/check_analytics_reachability.py:171-181 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (11 lines × 2) .github/scripts/check_backend_deploy_source_admission.py:70— .github/scripts/check_backend_deploy_source_admission.py:70-80 | .github/scripts/check_release_eligibility.py:23-33 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (11 lines × 2) .github/scripts/check_backend_deploy_source_admission.py:87— .github/scripts/check_backend_deploy_source_admission.py:87-97 | .github/scripts/check_release_eligibility.py:40-50 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (11 lines × 2) .github/scripts/check_backend_deploy_source_admission.py:151— .github/scripts/check_backend_deploy_source_admission.py:151-161 | .github/scripts/check_backend_deploy_source_admission.py:409-419 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (11 lines × 2) backend/database/action_items.py:1055— backend/database/action_items.py:1055-1065 | backend/database/conversations.py:1501-1511 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once.
Duplicated block (11 lines × 2) backend/database/announcements.py:101— backend/database/announcements.py:101-111 | backend/database/announcements.py:152-162 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (11 lines × 2) backend/database/conversations.py:94— backend/database/conversations.py:94-104 | backend/migrations/002_populate_historical_usage.py:51-61 — the copies span different directories, so extracting a shared function means choosing where it lives: put it somewhere both call sites can already reach — a location they all depend on today, or a new shared one if there is none — and call it from each site; until then, every change has to be made twice. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (11 lines × 2) backend/database/task_recommendations.py:189— backend/database/task_recommendations.py:189-199 | backend/database/task_recommendations.py:724-734 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (11 lines × 2) backend/database/workstreams.py:552— backend/database/workstreams.py:552-562 | backend/database/workstreams.py:680-690 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/database/workstreams.py:552` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (11 lines × 2) backend/dependencies.py:59— backend/dependencies.py:59-69 | backend/dependencies.py:93-103 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (11 lines × 2) backend/dependencies.py:105— backend/dependencies.py:105-115 | backend/dependencies.py:202-213 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/dependencies.py:105` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (11 lines × 2) backend/llm_gateway/gateway/executor.py:385— backend/llm_gateway/gateway/executor.py:385-395 | backend/llm_gateway/gateway/executor.py:398-408 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (11 lines × 2) backend/parakeet/transcribe.py:153— backend/parakeet/transcribe.py:153-163 | backend/parakeet/transcribe.py:236-246 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (11 lines × 2) backend/routers/action_items.py:739— backend/routers/action_items.py:739-749 | backend/routers/action_items.py:772-782 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/routers/action_items.py:739` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (11 lines × 2) backend/routers/apps.py:2040— backend/routers/apps.py:2040-2051 | backend/utils/mcp_client.py:674-684 — the copies span different directories, so extracting a shared function means choosing where it lives: put it somewhere both call sites can already reach — a location they all depend on today, or a new shared one if there is none — and call it from each site; until then, every change has to be made twice. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (11 lines × 2) backend/routers/developer.py:1448— backend/routers/developer.py:1448-1458 | backend/routers/mcp.py:495-505 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. Read the line range as the matched WINDOW rather than a finished unit: at `backend/routers/developer.py:1448` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (11 lines × 2) backend/routers/listen/parity_capture.py:33— backend/routers/listen/parity_capture.py:33-43 | backend/testing/parity_pack_v0/live_capture.py:36-46 — the copies span different directories, so extracting a shared function means choosing where it lives: put it somewhere both call sites can already reach — a location they all depend on today, or a new shared one if there is none — and call it from each site; until then, every change has to be made twice. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (11 lines × 2) backend/routers/mcp.py:378— backend/routers/mcp.py:378-388 | backend/routers/mcp_sse.py:852-862 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. Read the line range as the matched WINDOW rather than a finished unit: at `backend/routers/mcp.py:378` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (11 lines × 2) backend/routers/staged_tasks.py:568— backend/routers/staged_tasks.py:568-578 | backend/routers/staged_tasks.py:650-660 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (11 lines × 2) backend/routers/task_integrations.py:289— backend/routers/task_integrations.py:289-299 | backend/routers/task_integrations.py:301-311 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/routers/task_integrations.py:289` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (11 lines × 2) backend/scripts/stt/m_benchmark_streaming.py:162— backend/scripts/stt/m_benchmark_streaming.py:162-172 | backend/scripts/stt/m_benchmark_streaming.py:209-219 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/stt/m_benchmark_streaming.py:162` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (11 lines × 2) backend/scripts/stt/m_benchmark_streaming.py:183— backend/scripts/stt/m_benchmark_streaming.py:183-193 | backend/scripts/stt/m_benchmark_streaming.py:232-242 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (11 lines × 2) backend/scripts/stt/q_modulate_debug.py:158— backend/scripts/stt/q_modulate_debug.py:158-168 | backend/scripts/stt/r_ab_modulate_compare.py:121-132 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (11 lines × 2) backend/scripts/stt/s_benchmark_prerecorded_api.py:187— backend/scripts/stt/s_benchmark_prerecorded_api.py:187-198 | backend/scripts/stt/s_benchmark_prerecorded_api.py:211-221 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (8 lines × 2) .github/scripts/check_release_eligibility.py:58— .github/scripts/check_release_eligibility.py:58-65 | .github/scripts/check_release_eligibility.py:78-85 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (8 lines × 2) backend/database/action_items.py:207— backend/database/action_items.py:207-214 | backend/database/action_items.py:311-318 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (8 lines × 2) backend/database/apps.py:385— backend/database/apps.py:385-392 | backend/database/apps.py:407-414 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (8 lines × 2) backend/database/chat_first_intents.py:219— backend/database/chat_first_intents.py:219-226 | backend/database/chat_first_intents.py:406-413 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/database/chat_first_intents.py:219` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (8 lines × 2) backend/database/conversation_finalization_jobs.py:1285— backend/database/conversation_finalization_jobs.py:1285-1292 | backend/database/conversation_finalization_jobs.py:1305-1318 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (8 lines × 2) backend/database/workstreams.py:860— backend/database/workstreams.py:860-867 | backend/database/workstreams.py:1299-1306 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (8 lines × 2) backend/database/workstreams.py:1232— backend/database/workstreams.py:1232-1239 | backend/database/workstreams.py:1305-1312 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (8 lines × 2) backend/models/memory_apply.py:303— backend/models/memory_apply.py:303-310 | backend/routers/mcp_sse.py:1678-1685 — the copies span different directories, so extracting a shared function means choosing where it lives: put it somewhere both call sites can already reach — a location they all depend on today, or a new shared one if there is none — and call it from each site; until then, every change has to be made twice.
Duplicated block (8 lines × 2) backend/routers/apps.py:443— backend/routers/apps.py:443-451 | backend/routers/apps.py:1129-1136 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (8 lines × 2) backend/routers/chat.py:487— backend/routers/chat.py:487-494 | backend/routers/chat.py:1483-1490 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (8 lines × 2) backend/routers/integration.py:382— backend/routers/integration.py:382-389 | backend/routers/integration.py:712-719 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/routers/integration.py:382` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (8 lines × 2) backend/routers/integration.py:480— backend/routers/integration.py:480-487 | backend/routers/integration.py:495-502 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/routers/integration.py:480` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (8 lines × 2) backend/scripts/backfill_developer_api_key_memory_grants.py:114— backend/scripts/backfill_developer_api_key_memory_grants.py:114-121 | backend/scripts/backfill_mcp_key_full_access.py:96-103 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once.
Duplicated block (8 lines × 2) backend/scripts/check_module_stub_pollution.py:152— backend/scripts/check_module_stub_pollution.py:152-159 | backend/scripts/scan_import_time_side_effects.py:305-312 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (8 lines × 2) backend/scripts/deploy_status_report.py:637— backend/scripts/deploy_status_report.py:637-644 | backend/scripts/repair_cloud_run_traffic.py:270-277 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (8 lines × 2) backend/scripts/generate_swift_openapi_types.py:718— backend/scripts/generate_swift_openapi_types.py:718-725 | backend/scripts/generate_ts_openapi_types.py:269-276 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (8 lines × 2) backend/scripts/runtime_env_validation/workflows.py:190— backend/scripts/runtime_env_validation/workflows.py:190-197 | backend/scripts/runtime_env_validation/workflows.py:208-215 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (8 lines × 2) backend/scripts/stt/n_benchmark_02_prerecorded.py:337— backend/scripts/stt/n_benchmark_02_prerecorded.py:337-344 | backend/scripts/stt/o_benchmark_02_streaming.py:356-363 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/stt/n_benchmark_02_prerecorded.py:337` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (8 lines × 2) backend/scripts/stt/t_benchmark_multilang.py:202— backend/scripts/stt/t_benchmark_multilang.py:202-209 | backend/scripts/stt/t_benchmark_multilang.py:313-320 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/stt/t_benchmark_multilang.py:202` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (8 lines × 2) backend/scripts/stt/v_benchmark_parakeet_streaming.py:71— backend/scripts/stt/v_benchmark_parakeet_streaming.py:71-78 | backend/scripts/stt/v_benchmark_parakeet_streaming.py:110-117 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (8 lines × 2) backend/testing/chaos-oom/pusher_fixed.py:398— backend/testing/chaos-oom/pusher_fixed.py:398-405 | backend/testing/chaos-oom/pusher_vuln.py:340-347 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (8 lines × 2) backend/testing/replay_harness_llm_gateway_fake_upstream/oracle.py:67— backend/testing/replay_harness_llm_gateway_fake_upstream/oracle.py:67-76 | backend/testing/replay_harness_llm_streaming_wire_fidelity/oracle.py:95-102 — the copies span different directories, so extracting a shared function means choosing where it lives: put it somewhere both call sites can already reach — a location they all depend on today, or a new shared one if there is none — and call it from each site; until then, every change has to be made twice.
Duplicated block (8 lines × 2) backend/testing/replay_harness_phase0a/scenario.py:131— backend/testing/replay_harness_phase0a/scenario.py:131-138 | backend/testing/sync_cloud_tasks_stack/run.py:461-468 — the copies span different directories, so extracting a shared function means choosing where it lives: put it somewhere both call sites can already reach — a location they all depend on today, or a new shared one if there is none — and call it from each site; until then, every change has to be made twice. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (8 lines × 2) backend/utils/app_integrations.py:314— backend/utils/app_integrations.py:314-321 | backend/utils/app_integrations.py:902-910 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (8 lines × 2) backend/utils/llm/followup.py:10— backend/utils/llm/followup.py:10-17 | backend/utils/llm/openglass.py:15-22 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (12 lines × 2) .github/scripts/check_backend_deploy_source_admission.py:410— .github/scripts/check_backend_deploy_source_admission.py:410-421 | .github/scripts/check_release_eligibility.py:103-114 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. Read the line range as the matched WINDOW rather than a finished unit: at `.github/scripts/check_backend_deploy_source_admission.py:410` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (12 lines × 2) .github/scripts/desktop_backend_candidate_probe.py:103— .github/scripts/desktop_backend_candidate_probe.py:103-114 | .github/scripts/desktop_backend_candidate_probe.py:133-145 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `.github/scripts/desktop_backend_candidate_probe.py:133` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (12 lines × 2) backend/database/announcements.py:22— backend/database/announcements.py:22-33 | backend/database/announcements.py:55-66 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (12 lines × 2) backend/database/chat.py:105— backend/database/chat.py:105-117 | backend/database/chat.py:129-140 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/database/chat.py:105` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (12 lines × 2) backend/database/mcp_oauth.py:564— backend/database/mcp_oauth.py:564-575 | backend/database/mcp_oauth.py:603-614 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/database/mcp_oauth.py:564` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (12 lines × 2) backend/database/workstreams.py:413— backend/database/workstreams.py:413-424 | backend/database/workstreams.py:542-554 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/database/workstreams.py:413` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (12 lines × 2) backend/parakeet/main.py:196— backend/parakeet/main.py:196-207 | backend/parakeet/main.py:252-263 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/parakeet/main.py:196` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (12 lines × 2) backend/parakeet/main.py:221— backend/parakeet/main.py:221-232 | backend/parakeet/main.py:287-299 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/parakeet/main.py:287` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (12 lines × 2) backend/routers/integrations.py:485— backend/routers/integrations.py:485-496 | backend/routers/task_integrations.py:661-672 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. Read the line range as the matched WINDOW rather than a finished unit: at `backend/routers/integrations.py:485` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (12 lines × 2) backend/routers/memory_product.py:115— backend/routers/memory_product.py:115-126 | backend/routers/memory_product.py:223-234 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/routers/memory_product.py:115` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (12 lines × 2) backend/routers/staged_tasks.py:546— backend/routers/staged_tasks.py:546-557 | backend/routers/staged_tasks.py:630-641 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/routers/staged_tasks.py:546` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (12 lines × 2) backend/routers/sync.py:1698— backend/routers/sync.py:1698-1709 | backend/routers/sync.py:1838-1849 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (12 lines × 2) backend/routers/task_integrations.py:413— backend/routers/task_integrations.py:413-424 | backend/routers/task_integrations.py:493-504 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/routers/task_integrations.py:413` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (12 lines × 2) backend/routers/task_integrations.py:536— backend/routers/task_integrations.py:536-547 | backend/routers/task_integrations.py:576-587 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (12 lines × 2) backend/scripts/route_policy_inventory.py:937— backend/scripts/route_policy_inventory.py:937-948 | backend/scripts/route_policy_inventory.py:959-970 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/route_policy_inventory.py:937` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (12 lines × 2) backend/scripts/stt/g_pyannote_api.py:14— backend/scripts/stt/g_pyannote_api.py:14-25 | backend/scripts/stt/h_brainstorming.py:88-99 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (12 lines × 2) backend/scripts/stt/l_benchmark_prerecorded.py:155— backend/scripts/stt/l_benchmark_prerecorded.py:155-166 | backend/scripts/stt/m_benchmark_streaming.py:247-258 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once.
Duplicated block (12 lines × 2) backend/scripts/stt/r_modulate_stability.py:155— backend/scripts/stt/r_modulate_stability.py:155-166 | backend/scripts/stt/r_silence_sweep.py:147-158 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (12 lines × 2) backend/scripts/users/chats_analysis.py:131— backend/scripts/users/chats_analysis.py:131-143 | backend/scripts/users/chats_analysis.py:182-193 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (12 lines × 2) backend/scripts/vector_repair_outbox_worker_entrypoint.py:122— backend/scripts/vector_repair_outbox_worker_entrypoint.py:122-133 | backend/scripts/vector_repair_outbox_worker_entrypoint.py:263-274 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/vector_repair_outbox_worker_entrypoint.py:122` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (12 lines × 2) backend/testing/chaos-oom/load_generator.py:58— backend/testing/chaos-oom/load_generator.py:58-70 | backend/testing/chaos-oom/load_generator.py:104-115 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (12 lines × 2) backend/utils/apps.py:276— backend/utils/apps.py:276-287 | backend/utils/apps.py:494-505 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (12 lines × 2) backend/utils/conversations/location.py:16— backend/utils/conversations/location.py:16-27 | backend/utils/conversations/location.py:83-94 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (5 lines × 2) .github/scripts/desktop_release_metadata.py:59— .github/scripts/desktop_release_metadata.py:59-63 | .github/scripts/desktop_release_metadata.py:67-71 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `.github/scripts/desktop_release_metadata.py:59` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (5 lines × 2) backend/agent-proxy/main.py:783— backend/agent-proxy/main.py:783-787 | backend/agent-proxy/main.py:862-866 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (5 lines × 2) backend/database/desktop_update_channels.py:321— backend/database/desktop_update_channels.py:321-325 | backend/database/desktop_update_channels.py:417-421 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/database/desktop_update_channels.py:321` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (5 lines × 2) backend/database/llm_usage.py:46— backend/database/llm_usage.py:46-50 | backend/database/llm_usage.py:155-159 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (5 lines × 2) backend/database/memory_ledger.py:306— backend/database/memory_ledger.py:306-310 | backend/database/memory_ledger.py:353-357 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (5 lines × 2) backend/database/redis_db.py:604— backend/database/redis_db.py:604-610 | backend/database/redis_db.py:673-677 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/database/redis_db.py:604` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (5 lines × 2) backend/database/sync_ledger.py:135— backend/database/sync_ledger.py:135-139 | backend/database/sync_ledger.py:202-206 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/database/sync_ledger.py:202` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (5 lines × 2) backend/database/sync_ledger.py:387— backend/database/sync_ledger.py:387-391 | backend/database/sync_ledger.py:483-487 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/database/sync_ledger.py:483` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (5 lines × 2) backend/database/vector_db.py:1200— backend/database/vector_db.py:1200-1204 | backend/database/vector_db.py:1226-1230 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (5 lines × 2) backend/scripts/benchmark_translation.py:354— backend/scripts/benchmark_translation.py:354-360 | backend/scripts/benchmark_translation.py:372-376 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/benchmark_translation.py:354` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (5 lines × 2) backend/scripts/benchmark_translation.py:526— backend/scripts/benchmark_translation.py:526-531 | backend/scripts/benchmark_translation.py:536-540 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/benchmark_translation.py:526` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (5 lines × 2) backend/scripts/check_app_client_openapi_compatibility.py:416— backend/scripts/check_app_client_openapi_compatibility.py:416-420 | backend/scripts/check_app_client_openapi_compatibility.py:422-426 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (5 lines × 2) backend/scripts/inventory_app_client_schemas.py:1126— backend/scripts/inventory_app_client_schemas.py:1126-1130 | backend/scripts/scan_async_blockers.py:777-781 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once.
Duplicated block (5 lines × 2) backend/scripts/render_backend_runtime_env.py:59— backend/scripts/render_backend_runtime_env.py:59-63 | backend/scripts/render_backend_runtime_env.py:72-76 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/render_backend_runtime_env.py:59` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (5 lines × 2) backend/scripts/runtime_env_validation/cloud_run.py:161— backend/scripts/runtime_env_validation/cloud_run.py:161-165 | backend/scripts/runtime_env_validation/manifest.py:199-203 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (5 lines × 2) backend/scripts/smoke_mcp_chatgpt_review.py:61— backend/scripts/smoke_mcp_chatgpt_review.py:61-65 | backend/scripts/smoke_mcp_chatgpt_review.py:78-82 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (5 lines × 2) backend/scripts/stt/n_benchmark_02_prerecorded.py:310— backend/scripts/stt/n_benchmark_02_prerecorded.py:310-314 | backend/scripts/stt/n_benchmark_02_prerecorded.py:318-322 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/stt/n_benchmark_02_prerecorded.py:310` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (5 lines × 2) backend/scripts/stt/u_benchmark_der.py:380— backend/scripts/stt/u_benchmark_der.py:380-384 | backend/scripts/stt/x_benchmark_parakeet_der.py:259-263 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (5 lines × 2) backend/testing/sync_cloud_tasks_stack/run.py:741— backend/testing/sync_cloud_tasks_stack/run.py:741-745 | backend/testing/sync_cloud_tasks_stack/run.py:762-767 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (5 lines × 2) backend/utils/listen_pusher_session.py:248— backend/utils/listen_pusher_session.py:248-252 | backend/utils/listen_pusher_session.py:353-357 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (5 lines × 2) backend/utils/llm/conversation_processing.py:1353— backend/utils/llm/conversation_processing.py:1353-1357 | backend/utils/llm/conversation_processing.py:1430-1434 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (5 lines × 2) backend/utils/llm/gateway_shadow.py:203— backend/utils/llm/gateway_shadow.py:203-207 | backend/utils/llm/gateway_shadow.py:213-217 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (13 lines × 2) backend/database/apps.py:419— backend/database/apps.py:419-431 | backend/database/apps.py:447-459 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (13 lines × 2) backend/database/goals.py:615— backend/database/goals.py:615-627 | backend/database/goals.py:665-680 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (13 lines × 2) backend/database/memory_outbox_worker.py:776— backend/database/memory_outbox_worker.py:776-788 | backend/jobs/short_term_lifecycle_worker.py:173-185 — the copies span different directories, so extracting a shared function means choosing where it lives: put it somewhere both call sites can already reach — a location they all depend on today, or a new shared one if there is none — and call it from each site; until then, every change has to be made twice. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (13 lines × 2) backend/database/notifications.py:328— backend/database/notifications.py:328-340 | backend/database/notifications.py:373-385 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (13 lines × 2) backend/database/task_recommendations.py:773— backend/database/task_recommendations.py:773-785 | backend/database/task_recommendations.py:880-892 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/database/task_recommendations.py:773` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (13 lines × 2) backend/jobs/short_term_lifecycle_worker.py:227— backend/jobs/short_term_lifecycle_worker.py:227-239 | backend/utils/memory/canonical_required_processing.py:444-460 — the copies span different directories, so extracting a shared function means choosing where it lives: put it somewhere both call sites can already reach — a location they all depend on today, or a new shared one if there is none — and call it from each site; until then, every change has to be made twice. Read the line range as the matched WINDOW rather than a finished unit: at `backend/jobs/short_term_lifecycle_worker.py:227` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (13 lines × 2) backend/llm_gateway/routers/openai_compatible.py:331— backend/llm_gateway/routers/openai_compatible.py:331-343 | backend/llm_gateway/routers/openai_compatible.py:384-396 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/llm_gateway/routers/openai_compatible.py:331` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (13 lines × 2) backend/routers/auth.py:860— backend/routers/auth.py:860-873 | backend/routers/auth.py:960-972 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/routers/auth.py:860` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (13 lines × 2) backend/routers/chat.py:1393— backend/routers/chat.py:1393-1405 | backend/routers/chat.py:1453-1465 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/routers/chat.py:1393` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (13 lines × 2) backend/routers/desktop_agent_vm.py:756— backend/routers/desktop_agent_vm.py:756-768 | backend/routers/desktop_agent_vm.py:797-809 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/routers/desktop_agent_vm.py:756` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (13 lines × 2) backend/routers/desktop_chat.py:1127— backend/routers/desktop_chat.py:1127-1139 | backend/routers/desktop_chat.py:1282-1296 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/routers/desktop_chat.py:1127` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (13 lines × 2) backend/scripts/check_module_stub_pollution.py:208— backend/scripts/check_module_stub_pollution.py:208-221 | backend/scripts/scan_import_time_side_effects.py:378-390 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (13 lines × 2) backend/scripts/export_openapi.py:485— backend/scripts/export_openapi.py:485-497 | backend/scripts/route_policy_inventory.py:794-806 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once.
Duplicated block (13 lines × 2) backend/scripts/rag/current.py:121— backend/scripts/rag/current.py:121-133 | backend/scripts/rag/current.py:196-208 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (13 lines × 2) backend/scripts/stt/l_benchmark_prerecorded.py:194— backend/scripts/stt/l_benchmark_prerecorded.py:194-206 | backend/scripts/stt/l_benchmark_prerecorded.py:211-223 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (13 lines × 2) backend/scripts/verify_pusher_rollout_budget.py:226— backend/scripts/verify_pusher_rollout_budget.py:226-238 | backend/scripts/verify_pusher_rollout_budget.py:250-262 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/verify_pusher_rollout_budget.py:226` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (13 lines × 2) backend/testing/chaos-oom/pusher_fixed.py:121— backend/testing/chaos-oom/pusher_fixed.py:121-133 | backend/testing/chaos-oom/pusher_vuln.py:97-109 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (13 lines × 2) backend/utils/conversations/location.py:44— backend/utils/conversations/location.py:44-56 | backend/utils/conversations/location.py:114-126 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (13 lines × 2) backend/utils/conversations/memories.py:142— backend/utils/conversations/memories.py:142-155 | backend/utils/conversations/memories.py:201-213 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/utils/conversations/memories.py:142` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (13 lines × 2) backend/utils/llm/knowledge_graph.py:122— backend/utils/llm/knowledge_graph.py:122-134 | backend/utils/llm/knowledge_graph.py:216-228 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (13 lines × 2) backend/utils/memory/canonical_memory_adapter.py:1882— backend/utils/memory/canonical_memory_adapter.py:1882-1894 | backend/utils/memory/canonical_memory_adapter.py:1941-1953 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
D38 · OSV Dependency Vulnerabilities· Medium CVE · ×13
Medium CVE: [GHSA redacted] sdks/react-native/package-lock.json— @octokit/plugin-paginate-rest 6.1.2: [GHSA redacted] — @octokit/plugin-paginate-rest is not declared in this repo's manifests: it is pulled in transitively, so upgrade the dependency that requires it (— 9.2.2 is a MAJOR ahead of the resolved 6.1.2, so an `overrides` pin would force a breaking version under a dependent written against 6.1.2; upgrading the declaring package is the remedy, and where no patched release exists in your major line, record the exposure instead).
Medium CVE: [GHSA redacted] sdks/react-native/package-lock.json— @octokit/request 6.2.8: [GHSA redacted] — @octokit/request is not declared in this repo's manifests: it is pulled in transitively, so upgrade the dependency that requires it (— 8.4.1 is a MAJOR ahead of the resolved 6.2.8, so an `overrides` pin would force a breaking version under a dependent written against 6.2.8; upgrading the declaring package is the remedy, and where no patched release exists in your major line, record the exposure instead).
Medium CVE: [GHSA redacted] sdks/react-native/package-lock.json— @octokit/request-error 3.0.3: [GHSA redacted] — @octokit/request-error is not declared in this repo's manifests: it is pulled in transitively, so upgrade the dependency that requires it (— 5.1.1 is a MAJOR ahead of the resolved 3.0.3, so an `overrides` pin would force a breaking version under a dependent written against 3.0.3; upgrading the declaring package is the remedy, and where no patched release exists in your major line, record the exposure instead).
Medium CVE: [GHSA redacted] package-lock.json— @opentelemetry/core 1.30.1: [GHSA redacted] — @opentelemetry/core is not declared in this repo's manifests: it is pulled in transitively, so upgrade the dependency that requires it (— 2.8.0 is a MAJOR ahead of the resolved 1.30.1, so an `overrides` pin would force a breaking version under a dependent written against 1.30.1; upgrading the declaring package is the remedy, and where no patched release exists in your major line, record the exposure instead).
Medium CVE: PYSEC-2026-2132 backend/modal/requirements.txt— click 8.1.7: PYSEC-2026-2132 — click is not declared by the manifests that govern this lockfile: it is pulled in transitively here (the 8.4.0 this repo declares for it belongs to a separate install root and does not reach this one), so upgrade the dependency that requires it (or raise its floor in your own manifest and re-resolve (`uv lock --upgrade-package click`)). (in 3 dependency files: backend/modal/requirements.txt, backend/openapi-requirements.txt, backend/requirements.txt)
Medium CVE: PYSEC-2026-2132 backend/diarizer/requirements.txt— click 8.3.1: PYSEC-2026-2132 — upgrade to 8.3.3
Medium CVE: PYSEC-2024-38 plugins/composio/requirements.txt— fastapi 0.104.1: PYSEC-2024-38 — upgrade to 0.109.1 (in 11 dependency files)
Medium CVE: GO-2026-5024 sdks/device/go/go.mod— golang.org/x/sys 0.11.0 (golang.org/x/sys/windows): GO-2026-5024 — golang.org/x/sys is not a DIRECT requirement of this module: go.mod records it as `// indirect`, pulled in transitively, so raise it in place (run `go -C sdks/device/go get golang.org/x/sys@v0.44.0`, which updates the require line sdks/device/go/go.mod already holds for it).
Medium CVE: [GHSA redacted] desktop/windows/pnpm-lock.yaml— hono 4.12.29: [GHSA redacted] — hono is not declared by the manifests that govern this lockfile: it is pulled in transitively here (the ~4.12.27 this repo declares for it belongs to a separate install root and does not reach this one), so upgrade the dependency that requires it (or pin hono to 4.12.34 with an `overrides` entry (`pnpm.overrides` for pnpm)).
Medium CVE: [GHSA redacted] package-lock.json— hono 4.12.32: [GHSA redacted] — this repo declares hono ~4.12.27, a range that ALREADY admits the fixed 4.12.34, so there is no manifest edit to make here. Re-resolve the lock so hono moves onto 4.12.34 or later; if the flagged 4.12.32 comes back, a dependency is pinning it — upgrade that dependent, or pin hono with an `overrides` entry (`pnpm.overrides` for pnpm) so only one copy resolves.
Medium CVE: [GHSA redacted] plugins/omi-ms365-app/requirements.txt— jinja2 3.1.4: [GHSA redacted] — upgrade to 3.1.6. This is 1 of 3 advisories with a published fix this scan raises against jinja2 3.1.4, and their fixed versions do not agree — anything below 3.1.6 still leaves at least one of them open. Take this package to 3.1.6 or later: that is the floor for the package, not this row's target alone. This one row stands for the 3 advisories this scan raises against jinja2 3.1.4: [GHSA redacted], [GHSA redacted], [GHSA redacted].
Medium CVE: [GHSA redacted] backend/modal/requirements.txt— langchain 1.2.10: [GHSA redacted] — langchain is not declared by the manifests that govern this lockfile: it is pulled in transitively here (the 1.3.1 this repo declares for it belongs to a separate install root and does not reach this one), so upgrade the dependency that requires it (or raise its floor in your own manifest and re-resolve (`uv lock --upgrade-package langchain`)). (in 3 dependency files: backend/modal/requirements.txt, backend/pusher/requirements.txt, backend/requirements.txt)
Medium CVE: [GHSA redacted] plugins/requirements.txt— langchain 1.3.1: [GHSA redacted] — upgrade to 1.3.9
Duplicated block (16 lines × 2) .github/scripts/desktop_backend_candidate_probe.py:265— .github/scripts/desktop_backend_candidate_probe.py:265-280 | .github/scripts/desktop_backend_candidate_probe.py:318-346 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (16 lines × 2) backend/database/recurrence_inbox.py:163— backend/database/recurrence_inbox.py:163-180 | backend/database/recurrence_inbox.py:199-214 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/database/recurrence_inbox.py:163` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (16 lines × 2) backend/routers/chat.py:513— backend/routers/chat.py:513-528 | backend/routers/chat.py:1509-1524 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (16 lines × 2) backend/routers/developer.py:633— backend/routers/developer.py:633-648 | backend/routers/developer.py:669-684 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (16 lines × 2) backend/routers/integration.py:397— backend/routers/integration.py:397-412 | backend/routers/integration.py:527-542 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/routers/integration.py:397` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (16 lines × 2) backend/routers/sync.py:512— backend/routers/sync.py:512-527 | backend/routers/sync.py:913-928 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (16 lines × 2) backend/routers/task_integrations.py:735— backend/routers/task_integrations.py:735-750 | backend/routers/task_integrations.py:875-890 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (16 lines × 2) backend/scripts/check_module_stub_pollution.py:269— backend/scripts/check_module_stub_pollution.py:269-284 | backend/scripts/scan_import_time_side_effects.py:437-452 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/check_module_stub_pollution.py:269` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (16 lines × 2) backend/scripts/stlt_lifecycle_proof.py:333— backend/scripts/stlt_lifecycle_proof.py:333-348 | backend/scripts/stlt_lifecycle_proof.py:384-399 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/stlt_lifecycle_proof.py:333` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (16 lines × 2) backend/testing/chaos-oom/pusher_fixed.py:167— backend/testing/chaos-oom/pusher_fixed.py:167-182 | backend/testing/chaos-oom/pusher_vuln.py:143-158 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (16 lines × 2) backend/testing/chaos-oom/pusher_fixed.py:251— backend/testing/chaos-oom/pusher_fixed.py:251-266 | backend/testing/chaos-oom/pusher_vuln.py:195-210 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. Read the line range as the matched WINDOW rather than a finished unit: at `backend/testing/chaos-oom/pusher_fixed.py:251` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (16 lines × 2) backend/utils/app_integrations.py:222— backend/utils/app_integrations.py:222-240 | backend/utils/app_integrations.py:824-839 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (16 lines × 2) backend/utils/llm/knowledge_graph.py:153— backend/utils/llm/knowledge_graph.py:153-168 | backend/utils/llm/knowledge_graph.py:249-264 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Medium IaC: DS-0013 backend/Dockerfile— 'RUN cd ...' to change directory
Medium IaC: KSV-0001 backend/charts/agent-vm-reaper/prod_agent_vm_reaper_cronjob.yaml— Can elevate its own privileges
Medium IaC: KSV-0012 backend/charts/agent-vm-reaper/prod_agent_vm_reaper_cronjob.yaml— Runs as root user One securityContext edit clears this facet's near-duplicate rules together: KSV-0012, KSV-0020, KSV-0021.
Medium IaC: KSV-0104 backend/charts/agent-vm-reaper/prod_agent_vm_reaper_cronjob.yaml— Seccomp policies disabled One securityContext edit clears this facet's near-duplicate rules together: KSV-0030, KSV-0104.
Medium IaC: KSV-0125 backend/charts/agent-vm-reaper/prod_agent_vm_reaper_cronjob.yaml— Restrict container images to trusted registries
Medium IaC: KSV-0001 backend/charts/backend-listen/dev_omi_backend_listen_cronjob.yaml— Can elevate its own privileges
Medium IaC: KSV-0012 backend/charts/backend-listen/dev_omi_backend_listen_cronjob.yaml— Runs as root user One securityContext edit clears this facet's near-duplicate rules together: KSV-0012, KSV-0020, KSV-0021.
Medium IaC: KSV-0104 backend/charts/backend-listen/dev_omi_backend_listen_cronjob.yaml— Seccomp policies disabled One securityContext edit clears this facet's near-duplicate rules together: KSV-0030, KSV-0104.
Medium IaC: KSV-0125 backend/charts/backend-listen/dev_omi_backend_listen_cronjob.yaml— Restrict container images to trusted registries
Medium IaC: KSV-0001 backend/charts/backend-listen/prod_omi_backend_listen_cronjob.yaml— Can elevate its own privileges
Medium IaC: KSV-0012 backend/charts/backend-listen/prod_omi_backend_listen_cronjob.yaml— Runs as root user One securityContext edit clears this facet's near-duplicate rules together: KSV-0012, KSV-0020, KSV-0021.
Duplicated block (14 lines × 2) backend/database/candidates.py:773— backend/database/candidates.py:773-786 | backend/database/candidates.py:830-843 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (14 lines × 2) backend/database/memories.py:409— backend/database/memories.py:409-422 | backend/database/memories.py:762-775 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (14 lines × 2) backend/llm_gateway/routers/openai_compatible.py:141— backend/llm_gateway/routers/openai_compatible.py:141-154 | backend/llm_gateway/routers/openai_compatible.py:199-212 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/llm_gateway/routers/openai_compatible.py:141` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (14 lines × 2) backend/scripts/app_key_memory_grant_assignment_readiness.py:299— backend/scripts/app_key_memory_grant_assignment_readiness.py:299-312 | backend/scripts/mcp_api_key_scope_readiness.py:241-254 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/app_key_memory_grant_assignment_readiness.py:299` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (14 lines × 2) backend/scripts/runtime_env_validation/workflows.py:475— backend/scripts/runtime_env_validation/workflows.py:475-488 | backend/scripts/runtime_env_validation/workflows.py:521-534 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/runtime_env_validation/workflows.py:475` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (14 lines × 2) backend/scripts/stt/u_benchmark_parakeet_prerecorded.py:116— backend/scripts/stt/u_benchmark_parakeet_prerecorded.py:116-129 | backend/scripts/stt/u_benchmark_parakeet_prerecorded.py:135-148 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (14 lines × 2) backend/scripts/stt/v_benchmark_parakeet_streaming.py:179— backend/scripts/stt/v_benchmark_parakeet_streaming.py:179-192 | backend/scripts/stt/v_benchmark_parakeet_streaming.py:197-210 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (14 lines × 2) backend/testing/chaos-oom/load_generator.py:25— backend/testing/chaos-oom/load_generator.py:25-38 | backend/testing/chaos-oom/load_generator.py:75-88 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (14 lines × 2) backend/utils/conversations/memory_extraction_telemetry.py:160— backend/utils/conversations/memory_extraction_telemetry.py:160-173 | backend/utils/integration_telemetry.py:175-188 — the copies span different directories, so extracting a shared function means choosing where it lives: put it somewhere both call sites can already reach — a location they all depend on today, or a new shared one if there is none — and call it from each site; until then, every change has to be made twice. Read the line range as the matched WINDOW rather than a finished unit: at `backend/utils/integration_telemetry.py:175` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (14 lines × 2) backend/utils/llm/knowledge_graph.py:176— backend/utils/llm/knowledge_graph.py:176-189 | backend/utils/llm/knowledge_graph.py:271-284 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/utils/llm/knowledge_graph.py:176` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (14 lines × 2) backend/utils/memory/canonical_required_processing.py:666— backend/utils/memory/canonical_required_processing.py:666-679 | backend/utils/memory/canonical_required_processing.py:744-757 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/utils/memory/canonical_required_processing.py:666` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Hotspot: desktop/macos/agent/src/index.ts desktop/macos/agent/src/index.ts— desktop/macos/agent/src/index.ts changed 95 times in last 90 days, max complexity 315. 8 of those changes were fix/bug commits, so the churn is repair rather than feature work. Before the next change lands here, cover the area it touches with tests, then split that area out of the file so the following change is smaller than this one — a file this often edited pays the complexity back every time.
Hotspot: desktop/macos/agent/src/runtime/control-tools.ts desktop/macos/agent/src/runtime/control-tools.ts— desktop/macos/agent/src/runtime/control-tools.ts changed 72 times in last 90 days, max complexity 135. 10 of those changes were fix/bug commits, so the churn is repair rather than feature work. Before the next change lands here, cover the area it touches with tests, then split that area out of the file so the following change is smaller than this one — a file this often edited pays the complexity back every time.
Hotspot: backend/routers/mcp_sse.py backend/routers/mcp_sse.py— backend/routers/mcp_sse.py changed 58 times in last 90 days, max complexity 122. 8 of those changes were fix/bug commits, so the churn is repair rather than feature work. Before the next change lands here, cover the area it touches with tests, then split that area out of the file so the following change is smaller than this one — a file this often edited pays the complexity back every time.
Hotspot: backend/routers/sync.py backend/routers/sync.py— backend/routers/sync.py changed 62 times in last 90 days, max complexity 72. 8 of those changes were fix/bug commits, so the churn is repair rather than feature work. Before the next change lands here, cover the area it touches with tests, then split that area out of the file so the following change is smaller than this one — a file this often edited pays the complexity back every time.
Hotspot: backend/utils/conversations/process_conversation.py backend/utils/conversations/process_conversation.py— backend/utils/conversations/process_conversation.py changed 66 times in last 90 days, max complexity 46. 11 of those changes were fix/bug commits, so the churn is repair rather than feature work. Before the next change lands here, cover the area it touches with tests, then split that area out of the file so the following change is smaller than this one — a file this often edited pays the complexity back every time.
Hotspot: .github/scripts/check-release-process-guards.py .github/scripts/check-release-process-guards.py— .github/scripts/check-release-process-guards.py changed 50 times in last 90 days, max complexity 49. 26 of those changes were fix/bug commits, so the churn is repair rather than feature work. Before the next change lands here, cover the area it touches with tests, then split that area out of the file so the following change is smaller than this one — a file this often edited pays the complexity back every time.
Hotspot: backend/utils/sync/pipeline.py backend/utils/sync/pipeline.py— backend/utils/sync/pipeline.py changed 22 times in last 90 days, max complexity 106. 10 of those changes were fix/bug commits, so the churn is repair rather than feature work. Before the next change lands here, cover the area it touches with tests, then split that area out of the file so the following change is smaller than this one — a file this often edited pays the complexity back every time.
Hotspot: desktop/windows/src/renderer/src/hooks/useChat.ts desktop/windows/src/renderer/src/hooks/useChat.ts— desktop/windows/src/renderer/src/hooks/useChat.ts changed 38 times in last 90 days, max complexity 61. 13 of those changes were fix/bug commits, so the churn is repair rather than feature work. Before the next change lands here, cover the area it touches with tests, then split that area out of the file so the following change is smaller than this one — a file this often edited pays the complexity back every time.
Hotspot: backend/routers/payment.py backend/routers/payment.py— backend/routers/payment.py changed 33 times in last 90 days, max complexity 65. Frequent change and high complexity in one file compound: schedule the next change to it to include carving out the part being edited, behind tests written first.
Hotspot: backend/routers/developer.py backend/routers/developer.py— backend/routers/developer.py changed 71 times in last 90 days, max complexity 30. 7 of those changes were fix/bug commits, so the churn is repair rather than feature work. Before the next change lands here, cover the area it touches with tests, then split that area out of the file so the following change is smaller than this one — a file this often edited pays the complexity back every time.
Change coupling: rewind.ts ↔ index.ts desktop/windows/src/main/ipc/rewind.ts— `desktop/windows/src/main/ipc/rewind.ts` and `desktop/windows/src/preload/index.ts` change together 69% of the time (9 of the 13 commits that touched whichever of the two files changed less often, counting a file under its earlier names as well) with no explicit dependency — the edge is real but nothing declares it. Read the pair before acting: if one registers itself into the other through a hook or an initialiser, the missing dependency is DELIBERATE — the registration is the link, and it is meant not to be an import — and the thing to add is a comment on each side naming the other, not a merge; if they simply belong together, co-locate them; if neither holds, the coupling is hidden and worth breaking.
Change coupling: RealtimeHubController.swift ↔ RealtimeHubTools.swift desktop/macos/Desktop/Sources/FloatingControlBar/RealtimeHubController.swift— `desktop/macos/Desktop/Sources/FloatingControlBar/RealtimeHubController.swift` and `desktop/macos/Desktop/Sources/FloatingControlBar/RealtimeHubTools.swift` change together 61% of the time (25 of the 41 commits that touched whichever of the two files changed less often, counting a file under its earlier names as well). They sit in the same directory, and in this ecosystem sibling files there normally share one namespace/package — so a direct reference between them needs no import and this pass cannot see whether one exists. Read the pair before acting: if one file only DECLARES what the other consumes (a constants/types file beside its user), the co-change is definitional and the question is whether the split earns its keep; if they duplicate structure, extract the common part into a shared function or type they both call; if neither holds, the coupling is hidden and worth breaking.
Change coupling: kernel.ts ↔ worker-pool.ts desktop/macos/agent/src/runtime/kernel.ts— `desktop/macos/agent/src/runtime/kernel.ts` and `desktop/macos/agent/src/runtime/worker-pool.ts` change together 60% of the time (6 of the 10 commits that touched whichever of the two files changed less often, counting a file under its earlier names as well) with no explicit dependency between them. They sit in the same directory, but in this ecosystem each file is its own module — a sibling reference still needs an import — so the missing import edge is real: the coupling runs through shared behaviour, not a declared dependency. If they duplicate structure, extract the common part into one unit; otherwise the coupling is hidden and worth breaking.
Change coupling: CalendarReaderService.swift ↔ GmailReaderService.swift desktop/macos/Desktop/Sources/CalendarReaderService.swift— `desktop/macos/Desktop/Sources/CalendarReaderService.swift` and `desktop/macos/Desktop/Sources/GmailReaderService.swift` change together 60% of the time (6 of the 10 commits that touched whichever of the two files changed less often, counting a file under its earlier names as well). They sit in the same directory, and in this ecosystem sibling files there normally share one namespace/package — so a direct reference between them needs no import and this pass cannot see whether one exists. Read the pair before acting: if one file only DECLARES what the other consumes (a constants/types file beside its user), the co-change is definitional and the question is whether the split earns its keep; if they duplicate structure, extract the common part into a shared function or type they both call; if neither holds, the coupling is hidden and worth breaking.
Change coupling: acp.ts ↔ local-subprocess.ts desktop/macos/agent/src/adapters/acp.ts— `desktop/macos/agent/src/adapters/acp.ts` and `desktop/macos/agent/src/adapters/local-subprocess.ts` change together 58% of the time (7 of the 12 commits that touched whichever of the two files changed less often, counting a file under its earlier names as well) with no explicit dependency between them. They sit in the same directory, but in this ecosystem each file is its own module — a sibling reference still needs an import — so the missing import edge is real: the coupling runs through shared behaviour, not a declared dependency. If they duplicate structure, extract the common part into one unit; otherwise the coupling is hidden and worth breaking.
Change coupling: APIClient.swift ↔ OnboardingMemoryLogImportService.swift desktop/macos/Desktop/Sources/APIClient.swift— `desktop/macos/Desktop/Sources/APIClient.swift` and `desktop/macos/Desktop/Sources/Onboarding/OnboardingMemoryLogImportService.swift` change together 55% of the time (6 of the 11 commits that touched whichever of the two files changed less often, counting a file under its earlier names as well). They sit in different directories, but in this ecosystem the namespace is declared in the FILE, not by the folder — so the two may well share one namespace and reference each other with no import for this pass to see. Read the pair before acting: if one derives from or overrides the other, the dependency is explicit in the type declaration and the co-change is definitional; if one registers itself into the other through a hook or an initialiser, the missing dependency is DELIBERATE and the thing to add is a comment saying so; if they simply belong together, co-locate them; if none of these holds, the coupling is hidden and worth breaking.
Change coupling: clients.py ↔ gateway_serving.py backend/utils/llm/clients.py— `backend/utils/llm/clients.py` and `backend/utils/llm/gateway_serving.py` change together 55% of the time (6 of the 11 commits that touched whichever of the two files changed less often, counting a file under its earlier names as well) with no explicit dependency between them. They sit in the same directory, but in this ecosystem each file is its own module — a sibling reference still needs an import — so the missing import edge is real: the coupling runs through shared behaviour, not a declared dependency. If they duplicate structure, extract the common part into one unit; otherwise the coupling is hidden and worth breaking.
Change coupling: omiListen.ts ↔ omiListenClient.ts desktop/windows/src/main/ipc/omiListen.ts— `desktop/windows/src/main/ipc/omiListen.ts` and `desktop/windows/src/renderer/src/lib/omiListenClient.ts` change together 55% of the time (6 of the 11 commits that touched whichever of the two files changed less often, counting a file under its earlier names as well) with no explicit dependency — the edge is real but nothing declares it. Read the pair before acting: if one registers itself into the other through a hook or an initialiser, the missing dependency is DELIBERATE — the registration is the link, and it is meant not to be an import — and the thing to add is a comment on each side naming the other, not a merge; if they simply belong together, co-locate them; if neither holds, the coupling is hidden and worth breaking.
Duplicated block (15 lines × 2) backend/database/users.py:1710— backend/database/users.py:1710-1724 | backend/database/users.py:1822-1836 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (15 lines × 2) backend/routers/auth.py:381— backend/routers/auth.py:381-395 | backend/routers/auth.py:467-484 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (15 lines × 2) backend/routers/auth.py:756— backend/routers/auth.py:756-770 | backend/routers/auth.py:782-796 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/routers/auth.py:756` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (15 lines × 2) backend/routers/pusher.py:612— backend/routers/pusher.py:612-626 | backend/routers/pusher.py:631-645 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/routers/pusher.py:612` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (15 lines × 2) backend/scripts/runtime_env_validation/cloud_run.py:212— backend/scripts/runtime_env_validation/cloud_run.py:212-226 | backend/scripts/runtime_env_validation/cloud_run.py:252-266 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/runtime_env_validation/cloud_run.py:212` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (15 lines × 2) backend/scripts/verify_pusher_rollout_budget.py:117— backend/scripts/verify_pusher_rollout_budget.py:117-131 | backend/scripts/verify_pusher_rollout_gate.py:135-149 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once.
Duplicated block (15 lines × 2) backend/testing/chaos-oom/pusher_fixed.py:294— backend/testing/chaos-oom/pusher_fixed.py:294-308 | backend/testing/chaos-oom/pusher_vuln.py:237-251 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once.
Duplicated block (15 lines × 2) backend/utils/memory/bulk_legacy_backfill.py:397— backend/utils/memory/bulk_legacy_backfill.py:397-411 | backend/utils/memory/bulk_legacy_backfill.py:435-449 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Duplicated block (17 lines × 2) backend/database/conversations.py:377— backend/database/conversations.py:377-395 | backend/database/conversations.py:453-469 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited.
Duplicated block (17 lines × 2) backend/routers/pusher.py:322— backend/routers/pusher.py:322-340 | backend/testing/chaos-oom/pusher_fixed.py:288-304 — the copies span different directories, so extracting a shared function means choosing where it lives: put it somewhere both call sites can already reach — a location they all depend on today, or a new shared one if there is none — and call it from each site; until then, every change has to be made twice. Read the line range as the matched WINDOW rather than a finished unit: at `backend/routers/pusher.py:322` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (17 lines × 2) backend/scripts/inventory_app_client_schemas.py:1029— backend/scripts/inventory_app_client_schemas.py:1029-1045 | backend/scripts/stlt_lifecycle_proof.py:71-87 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/inventory_app_client_schemas.py:1029` it begins part-way through the construct above it, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that. Note that the copies do not run to the end of the range shown: their LAST lines are different code, not the same code under different names — the matched region ends inside that line. Extract the lines above it, and read the last line of each site separately.
Duplicated block (17 lines × 2) backend/scripts/stt/m_benchmark_streaming.py:287— backend/scripts/stt/m_benchmark_streaming.py:287-303 | backend/scripts/stt/m_benchmark_streaming.py:324-340 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/stt/m_benchmark_streaming.py:287` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (17 lines × 2) backend/scripts/stt/n_benchmark_02_prerecorded.py:200— backend/scripts/stt/n_benchmark_02_prerecorded.py:200-216 | backend/scripts/stt/n_benchmark_02_prerecorded.py:232-248 — both copies are in the same file, so extract the block into one function there and call it from each site — the copies drift apart the first time only one of them is edited. Read the line range as the matched WINDOW rather than a finished unit: at `backend/scripts/stt/n_benchmark_02_prerecorded.py:200` it does not close everything it opens, so those exact lines cannot be lifted as they stand — widen the region to the smallest complete statement or declaration that contains it, and extract that.
Duplicated block (17 lines × 2) backend/scripts/stt/r_modulate_stability.py:187— backend/scripts/stt/r_modulate_stability.py:187-203 | backend/scripts/stt/r_silence_sweep.py:181-198 — the copies sit in sibling files of one directory: extract the block into a single shared function in that directory and call it from each site, so a change lands once. The matched lines also transfer control out of the body holding them, which cannot survive a move into a called unit unchanged: have the extracted unit return that decision and let each site act on it.
Machine-readable and reproducible from this commit + frozen rubric — drop them straight into a contract appendix, a CRA dossier, or a downstream SCA / VEX tool.