Code Review: Current Or Branch Diff Skill

Codex compatibility note:

Invoke repository skills with $skill-name in Codex; this mirrored copy rewrites legacy Claude /skill-name references.

Task tracker mandate: BEFORE executing any workflow or skill step, create/update task tracking for all steps and keep it synchronized as progress changes.

User-question prompts mean to ask the user directly in Codex.

Ignore Claude-specific mode-switch instructions when they appear.

Strict execution contract: when a user explicitly invokes a skill, execute that skill protocol as written.

Subagent authorization: when a skill is user-invoked or AI-detected and its protocol requires subagents, that skill activation authorizes use of the required spawn_agent subagent(s) for that task.

Do not skip, reorder, or merge protocol steps unless the user explicitly approves the deviation first.

For workflow skills, execute each listed child-skill step explicitly and report step-by-step evidence.

If a required step/tool cannot run in this environment, stop and ask the user before adapting.

Codex Project-Reference Loading (No Hooks)

Codex uses static project-reference loading instead of runtime-injected project docs. When coding, planning, debugging, testing, or reviewing, open project docs explicitly using this routing.

Always read:

docs/project-config.json (project-specific paths, commands, modules, and workflow/test settings)
docs/project-reference/docs-index-reference.md (routes to the full docs/project-reference/* catalog)
docs/project-reference/lessons.md (always-on guardrails and anti-patterns)

Missing/stale context route: If docs/project-config.json, the docs index, lessons.md, CLAUDE.md, AGENTS.md, or any task-required reference doc is missing or stale, auto-run $project-init or the narrow setup route ($project-config, $docs-init, $scan-all, $scan --target=<key>, $claude-md-init) before ordinary project-specific work. If Codex mirrors or AGENTS.md are missing/stale, ask the user to run $sync-codex; do not auto-run it.

Situation-based docs:

Backend/CQRS/API/domain/entity changes: backend-patterns-reference.md, domain-entities-reference.md, project-structure-reference.md
Frontend/UI/styling/design-system: frontend-patterns-reference.md, scss-styling-guide.md, design-system/README.md
Spec authoring, docs/specs/ pathing, or TC format: feature-spec-reference.md, spec-system-reference.md, spec-principles.md
Behavior/public-contract changes or spec-test-code sync: workflow-spec-test-code-cycle-reference.md plus the spec docs above
Derived spec indexes/ERDs/reimplementation guides: spec-system-reference.md and source Feature Specs under docs/specs/
Integration test implementation/review: integration-test-reference.md
E2E test implementation/review: e2e-test-reference.md
Code review/audit work: code-review-rules.md plus domain docs above based on changed files

Do not read all docs blindly. Start from docs-index-reference.md, then open only relevant files for the task.

[BLOCKING] Execute skill steps in declared order. NEVER skip, reorder, or merge steps without explicit user approval. [BLOCKING] Before each step or sub-skill call, update task tracking: set in_progress when step starts, set completed when step ends. [BLOCKING] Every completed/skipped step MUST include brief evidence or explicit skip reason. [BLOCKING] If Task tools are unavailable, create and maintain an equivalent step-by-step plan tracker with the same status transitions.

[TOP REMINDER — WHY-REVIEW FINDINGS-VALIDATION GATE IS NON-NEGOTIABLE]

If this review produces ANY finding (Critical / High / Medium / Low) in standalone mode, you MUST invoke the $why-review skill through the skill invocation with --validate-findings <report-path> before any fix, docs-update, commit, or handoff. An actual skill call is the ONLY way to pass this gate — re-reading the cited file:lines yourself, "self-validating," or any inline/manual substitute does NOT count.

Create the todo the moment the first finding is recorded — never rely on memory: call task tracking → [Review Phase 6] Why-review findings validation gate — invoke $why-review so the gate is tracked and cannot be skipped. Inside $workflow-review-changes, stop after the report; parent step 2 owns validation. Full protocol in Phase 6; mirrored in the Closing Reminders at the bottom.

Quick Summary

Goal: Review current working-tree, staged, branch, or commit diffs across code, docs, config, infra, and non-code artifacts — finding correctness bugs, flaws, missing updates, stale docs, and convention drift with evidence — so every reviewed change is defect-free, evidence-backed, convention-aligned, and synchronized with required tests/docs before handoff; when code files changed, also prove the code stays easy to change.

Summary: read-this-if-nothing-else digest —

Report-driven and evidence-gated. Every finding is written to plans/reports/code-review-{date}-{slug}.md with file:line proof; speculation is forbidden, "looks fine" is not a verdict, codebase convention (grep 3+ examples) wins over textbook rules.
Findings are never auto-fixed on sight. Standalone mode runs validate (Phase 6 $why-review --validate-findings, an actual skill call) → fix (Phase 7) → restart $review-changes from Phase 0 over the full diff, looping until one whole pass has zero findings. Inside $workflow-review-changes you stop after the report and hand findings to the parent.
When code changed, three delegated gates are MANDATORY: Phase 3.5 $code-simplifier (clarity/maintainability), Phase 3.7 $integration-test-review Gate-7 coverage (every behavior change → covering test + spec TC), and — for every behavior change — Spec Drift Adjudication + the Dual-Feedback Ledger (the gap feeds BOTH spec AND tests).
Docs-update is the unconditional terminal step. Once the loop converges clean, Phase 8 $docs-update ALWAYS runs over the full changeset (deferred only to the parent inside the workflow).

Routing boundary: This skill reviews a git diff — working-tree (default), staged, branch, or commit. For an explicit file-set or SHA-range review, processing received review feedback, or a pre-completion verification gate over already-known scope, use code-review instead.

Shared engine (keep in sync): review-changes and code-review share the same review-protocol SYNC: blocks. Canonical source: .claude/skills/shared/sync-inline-versions.md; policy: SYNC:shared-protocol-duplication-policy. When you change a shared block in one skill, update the canonical file AND the sibling skill so the two never drift. The skills differ only in entry intent (diff vs explicit scope) and diff-specific gates (integration-test-sync, translation-sync, the Phase 3.7 integration-test-review coverage gate) — not in review quality.

Workflow:

Phase 0: Blast Radius — Call $graph-blast-radius skill FIRST (if .code-graph/graph.db exists)
Phase 0.3: Change Types — Detect high-risk change types; create risk tasks
Phase 0.5: Plan Compliance — Verify against active plan (conditional)
Phase 0.7: Surface Detection — AI categorizes changed files; creates dimension tasks
Phase 1: Collect — Run git status/diff, create report file
Phase 2: File Review — Review each changed file, update report incrementally
Phase 3: Fresh-Context Gate — Skip when findings already exist; run a second-round sub-agent only for an explicit user/workflow/high-risk synthesis trigger
Phase 3.5: Code-Simplifier Optimization (MANDATORY when code files changed) — Invoke $code-simplifier scoped to the changed code files to surface clarity/consistency/maintainability simplifications; record them as findings that flow into the same validation/fix loop (skip docs-only diffs)
Phase 3.7: Integration-Test-Review Coverage Gate (MANDATORY when behavior-bearing code changed) — Invoke $integration-test-review over the full diff; its 7 quality gates audit changed tests AND its Gate 7 (Change Coverage) maps every behavior-changing production file to a covering test (integration-first; unit fallback needs justification) and a spec TC. GAP/SPEC-GAP results become findings for the same validation/fix loop (skip docs-only diffs; deferred to the parent's dedicated step inside $workflow-review-changes)
Phase 4: Finalize — Generate critical issues, recommendations, suggested commit message
Phase 5: Docs Triage — Record stale-doc findings for validation/fix loop
Phase 6: Why-Review Findings Validation (standalone-only; REQUIRED before any standalone fix) — Whenever the report contains one or more findings, you MUST invoke the $why-review skill (an actual skill invocation-tool call) with --validate-findings to verify every finding is correct, proof-backed, reasonable, and best-practice before fixing. This is a genuine skill invocation — re-reading the cited lines yourself, "self-validating," or any inline/manual substitute does NOT satisfy this gate. When this skill is step 1 inside $workflow-review-changes, stop after the report; parent step 2 owns findings validation.
Phase 7: Recursive Fix + Full Re-Review Loop (standalone-only) — If validated findings remain in standalone mode, auto-fix them, then re-invoke $review-changes from Phase 0 with a fresh task breakdown over the full current diff; repeat until an entire review pass has zero findings. When inside $workflow-review-changes, parent steps 10-15 own plan/feature-implement/restart.
Phase 8: Mandatory Final Docs-Update Gate (MANDATORY — runs once the review/fix loop converges clean) — After the review reaches zero findings and all fixes are applied, ALWAYS invoke $docs-update over the full changeset as the terminal step so no stale docs survive. This is unconditional (not gated on a flagged finding) — $docs-update independently detects impacted docs the review may not have surfaced. When inside $workflow-review-changes, the parent workflow's $docs-update step owns this; do not run it locally.

Key Rules:

Report-driven: ALWAYS write findings to plans/reports/code-review-{date}-{slug}.md
MUST ATTENTION create todo tasks for ALL phases before starting
Skeptical: every claim needs file:line proof
Verify convention by grepping 3+ existing examples before flagging violations
Actively check DRY violations, YAGNI/KISS over-engineering, correctness bugs
When changed files include source code, run the Easy-to-Change gate: estimate future edit sites, coupling, hidden state, duplicated knowledge, unclear intent, and abstraction boundary health
When changed files include source code, run the Phase 3.5 $code-simplifier optimization gate over the changed code files — its simplification opportunities are findings that flow through the same Phase 6 validation → Phase 7 fix loop (never auto-applied unvalidated)
When changed files include behavior-bearing code, run the Phase 3.7 $integration-test-review coverage gate over the full diff — every behavior change must map to a covering test (integration-first) and a spec TC; GAP/SPEC-GAP verdicts are findings for the same Phase 6 → Phase 7 loop, never silently logged
Cross-reference changed files against related docs — flag stale docs, test specs, READMEs
MANDATORY FINAL step: once the review/fix loop converges to zero findings, ALWAYS run the Phase 8 $docs-update sweep over the full changeset — unconditional, never skipped on a clean verdict — why: a clean code review still leaves docs stale unless docs-update reconciles them against the actual changes
Findings are not eligible for auto-fix until Phase 6 why-review validation returns CLEAN for the current finding set
Every fix cycle invalidates the prior review result; restart $review-changes from Phase 0 and review the full updated diff, including the fixes
Continue review → validate findings → fix → full re-review until a complete review pass returns zero findings; do not add a fresh-context pass just because findings exist or a fix cycle restarted the review

MANDATORY Plan ToDo Task to discover and READ project-specific reference docs:

Search for code standards docs: *code-review*, *patterns*, *conventions*, *style-guide* — read any found

Search for architecture docs: *architecture*, *adr-*, README.md at service/module roots

Look for docs referencing changed technology areas (backend, frontend, infra, etc.)

Read docs most relevant to the categories of files changed

Prerequisites: MUST ATTENTION READ before executing:

Critical Purpose: Ensure quality — no flaws, no bugs, no missing updates, no stale content. Verify both artifacts AND documentation.

External Memory: For complex or lengthy work (research, analysis, scan, review), write intermediate findings and final results to a report file in plans/reports/ — prevents context loss and serves as deliverable.

Evidence Gate: MANDATORY — every claim, finding, and recommendation requires file:line proof or traced evidence with confidence percentage (>80% to act, <80% must verify first).

OOP & DRY Enforcement: MANDATORY — flag duplicated patterns that should be extracted to a base class, generic, or helper. Classes in the same group or suffix MUST ATTENTION inherit a common base (even if empty now — enables future shared logic and child overrides). Verify project has code linting/analyzer configured for the stack.

Code Review: Current Or Branch Diff

Review current changes or explicit branch/commit diffs against project standards.

Review Scope

Target: current working-tree changes by default; explicit branch/tag/commit diff when user asks branch comparison.

Use these sources:

Current changes: git status, git diff, and git diff --cached
Branch diff: git diff <base>...<head> plus git diff --name-only <base>...<head>
Commit range: git diff <base>..<head> plus git diff --name-only <base>..<head>

Review Mindset (NON-NEGOTIABLE)

Be skeptical. Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence >80%.

Verify correctness by reading actual implementations, never accept it at face value
Every finding MUST include file:line evidence (grep results, read confirmations)
Include a claim only when a trace proves it; otherwise leave it out of the report
Question assumptions: "Does this actually work?" → trace call path to confirm
Challenge completeness: "Is this all?" → grep related usages
Verify side effects: "What else does this change break?" → check consumers and dependents
No "looks fine" without proof — state what was verified and how

First Principle — Easy to Change

Apply this gate when diff includes source-code or code-adjacent files (.cs, .ts, .html, .scss, .css, tests, scripts, build/config-as-code). Pure docs-only changes skip this gate except for executable examples or code snippets.

Success metric: future change cost. DRY, SRP, abstraction, design patterns, naming, layering, tests — all serve one goal: make next change cheaper.

When evaluating code, refactor, test, or abstraction, ask: does this make next change cheaper or more expensive?

Reject "best practices" raising change cost: premature abstraction, speculative generality, leaky indirection, ceremony without payoff.
Name real enemies in findings: coupling, hidden state, duplicated knowledge, unclear intent, irreversible decisions exposed too early.
Favor project-owned boundaries around external libraries, e.g. component/service input-output contracts, when they localize future library changes; reject pass-through wrappers adding ceremony without lowering change cost.
Simpler design easy to change beats sophisticated design that isn't.

Apply this lens before specific rules, patterns, or checklists below. If downstream rule raises change cost, this principle wins.

Core Principles (ENFORCE ALL)

YAGNI — Flag code solving hypothetical future problems (unused parameters, speculative interfaces, premature abstractions) KISS — Flag unnecessarily complex solutions. "Is there a simpler way meeting same requirement?" DRY — Actively grep for similar/duplicate code before accepting new code. 3+ similar patterns → flag for extraction. Clean Code — Readable > clever. Names reveal intent. Functions do one thing. No deep nesting. Follow Convention — Before flagging ANY pattern violation, grep for 3+ existing examples. Codebase convention wins over textbook rules. No Flaws/No Bugs — Trace logic paths. Verify edge cases (null, empty, boundary values). Check error handling covers failure modes. Proof Required — Every claim backed by file:line evidence or grep results. Speculation FORBIDDEN. Doc Staleness — Cross-reference changed files against related docs (feature docs, test specs, READMEs). Flag stale or missing updates.

Run python .claude/scripts/code_graph batch-query <f1> <f2> --json on changed files for test coverage and caller impact.

Blast Radius Pre-Analysis (MANDATORY FIRST STEP)

IMPORTANT MANDATORY MUST ATTENTION: FIRST action in every review. Call $graph-blast-radius BEFORE any other review work.

If .code-graph/graph.db exists, run graph-blast-radius analysis before reviewing changes:

Call $graph-blast-radius skill (runs python .claude/scripts/code_graph blast-radius --json)
Include in review: impacted files count, untested changes, risk level based on blast radius size
Use results to prioritize file review order (highest-impact files first)

Graph-Assisted Change Review

For each changed file, trace full impact:

python .claude/scripts/code_graph trace <changed-file> --direction downstream --json — all files affected by changes
Flag any affected file NOT covered by tests
Catches cross-service impact simple diff review misses

Review Approach (Report-Driven Multi-Phase — CRITICAL)

MANDATORY FIRST: Create Todo Tasks for Review Phases Before starting, call task tracking with:

[ ] [Review Phase 0] Run $graph-blast-radius to analyze change impact - in_progress (MUST ATTENTION BE FIRST)
[ ] [Review Phase 0.3] Detect high-risk change types, create risk tasks - pending
[ ] [Review Phase 0.7] Categorize changed files, create dimension review tasks - pending
[ ] [Review Phase 0.5] Plan compliance check (skip if no active plan) - pending
[ ] [Review Phase 1] Get changes and create report file - pending
[ ] [Review Phase 2] Review file-by-file and update report - pending
[ ] [Review Phase 3] Evaluate fresh-context gate; skip when findings already exist - pending
[ ] [Review Phase 3.5] Run $code-simplifier on changed code files to optimize code quality - pending (MANDATORY when code files changed; skip docs-only diffs)
[ ] [Review Phase 3.7] Run $integration-test-review coverage gate over full diff - pending (MANDATORY when behavior-bearing code changed; skip docs-only diffs; deferred to parent step inside $workflow-review-changes)
[ ] [Review Phase 4] Generate final review findings - pending
[ ] [Review Phase 5] Record stale-doc findings for validation/fix loop - pending
[ ] [Review Phase 6] Why-review findings validation gate before any fix - pending (MANDATORY when findings exist)
[ ] [Review Phase 7] Auto-fix validated findings and restart $review-changes from Phase 0 - pending (MANDATORY when validated findings remain)
[ ] [Review Phase 8] Run $docs-update over full changeset to sync all impacted docs - pending (MANDATORY FINAL — always runs once review converges to zero findings; never skipped)

Update todo status as each phase completes.

Note: If Phase 1 reveals 10+ changed files, replace Phase 2-4 tasks with Systematic Review Protocol tasks: [Review Phase 2] Categorize and fire parallel sub-agents, [Review Phase 3] Synchronize and cross-reference, [Review Phase 3.5] Run $code-simplifier on changed code files, [Review Phase 3.7] Run $integration-test-review coverage gate, [Review Phase 4] Generate consolidated report

Phase 0: Run Graph Blast Radius Analysis (MANDATORY FIRST STEP)

IMPORTANT MANDATORY MUST ATTENTION: FIRST action before ANY other review work.

Call $graph-blast-radius skill
Record in report: changed files count, impacted files count, untested changes, risk level
Use blast radius output to prioritize which files to review most carefully in Phase 2
If .code-graph/graph.db does not exist, note "Graph not available — skipping blast radius" and proceed to Phase 0.3

Phase 0.3: Change Type Detection + Risk Tasks (MANDATORY)

Purpose: Identify HIGH-RISK change types in this diff before dimensional review. Each detected type creates a focused risk task. Change types are ORTHOGONAL to file category: the same file can be both a migration AND a security change — detect all independently.

Step 1: Detect change types

git diff --name-only HEAD       # unstaged
git diff --cached --name-only   # staged
# For branch or commit-range review, use the user-provided diff source:
git diff --name-only <base>...<head>

Evaluate each change type for this diff:

| Change Type | Detection Signal (adapt to project's actual conventions) | TRUE if... | | ------------------ | ----------------------------------------------------------------------------------------------------------------------------------- | --------------------------------------------------------- | | DepUpgrade | Dependency manifest changed (package.json, *.csproj, Gemfile, go.mod, requirements.txt, Cargo.toml, pom.xml, etc.) | A version number changed in any dependency manifest | | Migration | File path or name suggests schema change (contains migration, schema, alter_table, or matches project's migration convention) | Any migration-convention file appears in the diff | | BusEvent | New or modified event/message definition or consumer (infer from project conventions: consumer naming, message type directories) | A consumer or event class is new or its contract changed | | ApiContract | API definition file changed (controller, route handler, OpenAPI/GraphQL schema) with route or field differences | Diff shows route/action/field additions or removals | | SecurityChange | Auth/permission definition changed — infer from project conventions (auth middleware, permission constants, policy definitions) | Any auth or permission gate is added, removed, or changed | | ConfigChange | Configuration files changed (e.g., *.json, *.yaml, *.env*, *Config*, *Options*, *Settings*, *.toml) | Any config-convention file appears | | InfraChange | Infrastructure definition changed (Dockerfile, docker-compose*.yml, CI/CD pipelines, k8s manifests, IaC files) | Any infra-convention file appears |

Record in report:

## Change Type Analysis
DepUpgrade: [YES/NO] | Migration: [YES/NO] | BusEvent: [YES/NO]
ApiContract: [YES/NO] | SecurityChange: [YES/NO] | ConfigChange: [YES/NO] | InfraChange: [YES/NO]

Step 2: Create change-type risk tasks (ALWAYS before any review work)

MANDATORY: Call task tracking for each TRUE signal. Do NOT create tasks for FALSE signals. The concerns listed are starting points — apply domain knowledge beyond them.

| Condition | task tracking subject | Key concerns to investigate (starting points — expand with domain knowledge) | | ------------------- | ---------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | DepUpgrade TRUE | [Review-DepUpgrade] Dependency upgrade — semver, breaking changes, security advisories | Major/minor/patch? Read upstream CHANGELOG for breaking API changes. Grep deprecated API usage. Check transitive dependency changes. Known security advisories for new version? Peer dependency compatibility? Tests still passing? | | Migration TRUE | [Review-Migration] DB migration — rollback path, volume impact, zero-downtime | Rollback/Down script exists? Table size estimate — large tables need lock analysis. NOT NULL column without default on non-empty table? Indexes created with no-lock option? Deployment ordering (before/after service deploy)? Backfill idempotent if run twice? | | BusEvent TRUE | [Review-BusEvent] Cross-service event/message — consumer, idempotency, retry, poison pill | Consumer exists for new event? Retry strategy: prerequisite data not synced → wait-retry vs silent skip? Handler safe to run twice (idempotency)? Malformed message handling / dead-letter configured? Ordering assumptions vs broker guarantees? | | ApiContract TRUE | [Review-ApiContract] API contract change — backward compat, client alignment, auth | Additive or breaking? Breaking → versioning or coordinated deploy required. All callers (UI, other services, tests) still compatible? New endpoint protected appropriately? No required response fields added without client update? | | SecurityChange TRUE | [Review-SecurityChange] Security/permission change — all paths covered, no privilege escalation | All code paths reaching the gate covered? Negative test verifying unauthorized access DENIED? Privilege escalation possible? BOTH enforcement AND display control updated? Permission definition in single authoritative place (no duplicated strings risking drift)? | | ConfigChange TRUE | [Review-ConfigChange] Config/env change — all environments, no secrets committed | New config key present in ALL environment configs? Hardcoded default masking missing production config? Any secret value in the diff? → CRITICAL if yes. Documented in setup guide? App fails fast if config missing? | | InfraChange TRUE | [Review-InfraChange] Infrastructure change — env parity, no dev values in prod, reproducible build | Change affects all environments consistently? Hardcoded dev values (localhost, debug flags, dev credentials)? Pinned image/dependency versions? Local dev impact documented? CI/CD secret/permission requirements documented? |

AI-SDD risk lenses: Apply these lenses when the changed files touch specs, workflows, tooling, or shared guidance.

| Lens | Review focus | | --------------------------- | ---------------------------------------------------------------------------------------------- | | Contract/API/routes | Public behavior, clients, generated specs, and regression tests still agree. | | Permissions/security-review | Enforcement, display controls, negative tests, and authoritative permission definitions align. | | Config/flags | All environments, examples, fail-fast behavior, and docs are current. | | Docs/spec/test drift | Canonical specs, Section 8 TCs, dashboards, and test code are synchronized or explicitly N/A. | | Generated mirrors | Shared skill/workflow/tooling changes were synced to generated agent surfaces. | | Reference-only artifacts | AI-extracted specs/TCs remain draft/reference until accepted by the owning review gate. |

Step 3: Work through change-type tasks before dimensional review

For each created change-type task:

Set task to in_progress
Work through ALL applicable concerns — the table above is a starting point, not a ceiling
For each concern: cite file:line for PASS or describe finding for FAIL/WARN
Write findings under ## {Task Subject} Findings in report
Set task to completed

IMPORTANT: Complete ALL change-type tasks FIRST, then proceed to Phase 0.7. If no change-type signals detected, log "No high-risk change types detected" and proceed.

Phase 0.7: Change Surface Detection + Dynamic Review Tasks (MANDATORY)

Purpose: Let AI categorize the changes by nature and create review tasks accordingly. Derive categories from what the project's actual changed files are, never assume a fixed set. Think, don't classify into a preset grid. The AI owns this step entirely.

Step 1: Derive categories from the diff

git diff --name-only HEAD        # unstaged
git diff --cached --name-only    # staged
# For branch or commit-range review, use the user-provided diff source:
git diff --name-only <base>...<head>

For each changed file, infer its category by examining:

Language/extension: What technology or domain does this file belong to?
Directory semantics: What layer, module, or concern does this path represent in the project?
Change nature: Is this logic, data schema, configuration, documentation, infrastructure, styling, testing, or tooling?

Do NOT map to fixed buckets. Derive categories that fit the current repository's actual structure and vocabulary.

Common category types to consider as starting points (not exhaustive — derive what fits):

Server-side logic — business rules, API handlers, services, consumers, event processors
Client-side logic — UI components, state management, API integration
Data/Schema — migrations, schemas, seed data, domain models
Styles/Assets — CSS/SCSS, design tokens, images, fonts
Configuration — app settings, env vars, feature flags
Infrastructure — Docker, CI/CD, pipelines, cloud manifests
Documentation/Specs — markdown docs, ADRs, feature specs, test specs
Tests — unit, integration, E2E test files
Build/Tooling — build scripts, linters, formatters, bundlers, agent scripts
Security — auth config, permission definitions, certificates

Record in report:

## Change Surface
{Category name} ({category type}): {N} files
{Category name} ({category type}): {M} files
...

Step 2: For each category, enumerate concerns and create a task

This is where you THINK, not fill in blanks. Apply SYNC:category-review-thinking for each category.

For EACH identified category:

Understand the domain: What is this category's purpose? What invariants govern it? Who depends on it?
Read project conventions: Grep for style guides, patterns docs, READMEs specific to this area
Derive concerns from first principles — DO NOT limit to any fixed list; trust your domain knowledge
Create a task tracking task named [Review-{Category}] {brief concern summary} listing derived concerns
Select the appropriate sub-agent type (see Sub-Agent Type Selection)

ALWAYS create: [Review-General] — universal quality: correctness, YAGNI/KISS/DRY, doc staleness, test coverage. Runs across ALL changed files regardless of other categories.

Sub-Agent Type Selection:

| Category Nature | agent_type | | -------------------------------------- | ------------------------------------------------------------- | | Code logic (any stack) | code-reviewer | | Implements a documented spec/PBI | spec-compliance-reviewer (pre-pass, before code-reviewer) | | Security, auth, permissions | security-auditor | | Performance, query efficiency, latency | performance-optimizer | | Documentation, plans, specs, ADRs | general-purpose | | Infrastructure, CI/CD, config | general-purpose | | Mixed or default | code-reviewer |

Spec-compliance pre-pass (when the changeset implements a documented docs/specs/**capability or a PBI/story): spawnspec-compliance-reviewerFIRST — it verifies the implementation matches the spec (catches spec drift, missing requirements, extra features) BEFORE thecode-reviewerquality pass runs. Skip when no spec/PBI governs the change (thencode-revieweris the sole code pass). This is the one wired dispatch site forspec-compliance-reviewer (sub-agent-selection-guide.md "Spec compliance" row).

UI/frontend dimension (OWNED by this skill): When a Client-side logic or Styles/Assets category surfaces frontend files matching the project's configured frontend/UI file patterns, $review-changes owns the UI review and invokes $review-ui as its UI dimension — preferably as a dedicated ui-ux-designer sub-agent spawned in the same parallel batch as the other dimensional agents (inline-fold its checklist only when sub-agent spawning is unavailable). The checklist: long-content overflow (wrap vs ellipsis+tooltip), responsive multi-screen via flex, flex-grow vs fixed sizing (prefer min/max + flex over fixed px), z-index scale discipline (no raw numbers, no !important), and SCSS/BEM quality. This is the SAME behavior in both standalone and workflow contexts — $review-ui is NOT a separate workflow step; it always runs here. Skip entirely if no frontend files changed.

Step 3: Work through tasks in order

For each created task:

Set task to in_progress before starting
Review ONLY files in that category's scope
Apply SYNC:category-review-thinking — trust your domain knowledge beyond the examples there
Write findings to report under ## {Task Subject} Findings section
Set task to completed before starting next task

NEVER mark a dimension task completed by scanning. Work through each relevant file explicitly. For large categories (10+ files): escalate to a parallel sub-agent using the Systematic Review Protocol.

Phase 0.5: Plan Compliance Check (CONDITIONAL — only when active plan exists)

Check ## Plan Context in injected context:

If "Plan: none" → skip, log "No active plan — skipping plan compliance"
If "Plan: {path}" → load plan and verify:

Read {plan-path}/plan.md — get phase list and scope
Read relevant phase files — extract files to modify, test specifications, success criteria
Verify (MUST ATTENTION — all four):
- Scope match — changed files listed in plan phases (warn on unplanned files)
- Test evidence — tests mapped to completed phases have evidence (file:line), not "TBD"
- Success criteria met — phase success criteria satisfied by changes
- Test intent traceability — mapped tests name the business rule/invariant they protect, not just current behavior
Add "Plan Compliance" section to review report

Phase 1: Get Changes and Create Report File (MUST ATTENTION)

Identify diff source: current working tree, staged changes, branch comparison, or commit range
Run git status for current changes, or git diff --name-only <base>...<head> for branch comparisons
Run git diff or git diff <base>...<head> to see actual changes
Create plans/reports/code-review-{date}-{slug}.md
Initialize with Scope, Files to Review, Blast Radius Summary sections

Phase 2: File-by-File Review (Build Report Incrementally)

For EACH changed file, read and immediately update report with:

File path and change type (added/modified/deleted)
Change Summary: what modified/added
Purpose: why change exists
Convention check: Grep 3+ similar patterns — does new code follow existing convention?
Correctness check: Trace logic paths — handles null, empty, boundary values, error cases?
DRY check: Grep similar/duplicate code — does this logic already exist elsewhere?
Intention check: Does change serve stated purpose? Flag unrelated modifications
Logic trace: Trace one happy path + one error path. Logic matches requirements?
Semantic correctness: Does the artifact DO what it's supposed to?
Issues Found: naming, typing, responsibility, patterns, bugs, over-engineering, logic errors
Continue to next file, repeat

Phase 3: Fresh-Context Gate (Conditional Protocol — branch on findings and Phase 0.7 surface)

Protocol: SYNC:double-round-trip-review + SYNC:fresh-context-review + SYNC:review-protocol-injection (all inlined above). INVARIANT: Phase 3 is review-only. It may add findings, but it MUST NOT fix or validate them. Existing findings do not require a fresh-context re-review; any non-zero finding set flows to Phase 6 why-review validation, then Phase 7 auto-fix + full $review-changes restart from Phase 0. A Phase 7 restart alone is NOT a Phase 3 trigger.

Entry gate:

If Phase 2 or any dimensional review already found findings, SKIP Phase 3. Record: Skipped fresh-context pass because findings already exist; Phase 6 why-review validation is the required next gate. Then proceed to Phase 4 consolidation and Phase 6 validation.
If there are zero findings and no explicit independent-review trigger, SKIP Phase 3. Record: Skipped fresh-context pass because the current review is clean and no second-round trigger exists. Then proceed to Phase 4 finalization.
Run Phase 3 only when the current finding set is zero and at least one trigger exists:
- the user explicitly requested a second-round/fresh-context review;
- the selected workflow explicitly requires an independent reviewer for this invocation;
- high-risk multi-domain changes need synthesis before a clean verdict.

Anti-waste rule: Do not run Phase 3 to re-review known findings before Phase 6. Do not run Phase 3 solely because Phase 7 restarted the review after fixes. The restarted review is already the required full pass; if it has zero findings and no explicit trigger above, finalize cleanly.

If the entry gate allows Phase 3, check categories from Phase 0.7 — if multiple distinct domains changed (e.g., server-side + client-side), run Synthesis Mode. Otherwise run Holistic Mode.

[SYNTHESIS MODE — when multiple distinct domains changed]

Spawn a Synthesis Agent as Round 2. Purpose: catch cross-boundary issues individual dimensional tasks cannot see.

When constructing Agent call prompt:

Copy Agent call shape from SYNC:review-protocol-injection template verbatim, agent_type: "code-reviewer"
Embed all 11 universal SYNC blocks verbatim

Set Task as:

Synthesis review — cross-boundary concerns ONLY across the changed domains in this diff.
You have these dimensional findings as context: {summary from each dimensional task}.
Re-read ALL changed files from scratch via your own tool calls.

Focus ONLY on cross-boundary concerns — do NOT re-review each domain's internals:
1. Contract Alignment: Do callers match what callees expose? (routes, parameters, field names, types)
2. Data Consistency: Are field names/types consistent across layer boundaries?
3. Security Boundary: Is auth enforced on BOTH sides (enforcement AND display control)?
4. Cross-Layer Naming: Same concept named differently across layers?
5. Missing Wiring: New producer with no consumer? New consumer with no producer? New feature with no doc?
6. Documentation: Docs reflect changes in BOTH domains together?

Set Target Files as "use the selected diff source from Phase 1"
Set report path as plans/reports/synthesis-review-{date}.md

After sub-agent returns:

Read synthesis report
Integrate findings as ## Synthesis Round Findings in main report — DO NOT filter or override
If findings exist: do NOT fix here; mark Phase 3 complete and proceed to Phase 6 why-review validation
If no findings exist: proceed to Phase 4 finalization as a clean synthesis pass

[HOLISTIC MODE — when single domain changed]

No cross-boundary synthesis needed. Spawn standard holistic Round 2.

When constructing Agent call prompt:

Copy Agent call shape from SYNC:review-protocol-injection template verbatim
Select agent_type based on domain's dominant concern (see Sub-Agent Type Selection)
Set Task as: "Review the selected diff holistically. Focus on big picture — overall technical approach coherence, architecture layers, logic placement (lowest layer), DRY violations, YAGNI/KISS, function complexity. Domain: {category from Phase 0.7} — apply domain knowledge for this category accordingly."
Set Target Files as "use the selected diff source from Phase 1"
Set report path as plans/reports/code-review-changes-round{N}-{date}.md

After sub-agent returns:

Read sub-agent's report
Integrate findings as ## Round {N} Findings (Fresh Sub-Agent) in main report — DO NOT filter or override
If findings exist: do NOT fix here; mark Phase 3 complete and proceed to Phase 6 why-review validation
If no findings exist: proceed to Phase 4 finalization as a clean holistic pass
Final verdict must incorporate findings from ALL review passes executed in this invocation

The following checks are handled by sub-agent but can be verified in Phase 4:

Clean Code & Over-engineering Checks:

MUST ATTENTION YAGNI: Code solving hypothetical future problems? Unused params, speculative interfaces?
MUST ATTENTION KISS: Unnecessarily complex solution? Could this be simpler while meeting the same requirement?
MUST ATTENTION Function complexity: Methods too long? Nesting too deep? Multiple responsibilities?
MUST ATTENTION Readability: Would a new team member understand without reading the full implementation?

Documentation Staleness Check (REQUIRED):

For each changed file, identify related documentation:

Search for feature docs, architecture references, READMEs at module/service roots, API docs, test specs, setup guides
Flag any doc where content no longer matches the changed artifact
Flag missing docs for new features or components that should be documented
Flag in the report with the specific stale section and what changed. Do not fix yet; Phase 6 must validate the finding before Phase 7 invokes $docs-update or applies doc edits.

Spec Drift Adjudication (REQUIRED when behavior changed): Apply SYNC:spec-drift-adjudication. For every behavior-bearing change, compare it against the canonical Feature Spec under docs/specs/ and classify any divergence as CODE-WRONG (change violates an intended spec rule/AC/invariant → BLOCKING finding, fix code/test), SPEC-STALE (intentional behavior change the spec no longer reflects → route to $spec [update] + $spec [mode=tests] [update]), AMBIGUOUS (a direct user question before editing either side), or SPEC-SILENT (code correctly enforces an invariant no spec artifact states → ENRICH: add the §4 BR/§3 AC + a §8 TC via $spec [update] + $spec [mode=tests], then a guarding test). Record the verdict per changed behavior (Spec in sync when no divergence). Do not normalize drift just because code/tests pass. This is the bidirectional generalization of the post-bugfix "Was spec wrong?" check — it runs for ALL behavior-changing reviews, not only post-bugfix. Flag findings here; Phase 6 validates and Phase 7 fixes (CODE-WRONG fixes route through the fix loop; SPEC-STALE and SPEC-SILENT fixes route to the canonical spec updater before $docs-update).

Correctness & Bug Detection: Apply SYNC:bug-detection — null safety, boundaries, error handling, resource cleanup, concurrency.

Test Spec Verification: Apply SYNC:test-spec-verification — locate specs, verify coverage, flag gaps.

Integration Test Sync: Apply SYNC:integration-test-sync-check — surface missing tests via a direct user question.

Translation Sync: Apply SYNC:translation-sync-check — for multilingual UI text changes, require translation updates or explicit user risk acceptance.

Phase 3.5: Code-Simplifier Quality Optimization (MANDATORY when code files changed)

Purpose: A correctness review proves the change WORKS; this gate proves the changed code stays easy to read, consistent, and cheap to change. Bug-finding (Phases 2-3) and simplification optimization are different lenses — run both. $code-simplifier is the canonical owner of clarity/consistency/maintainability refinement, so this skill delegates to it rather than duplicating that logic.

Entry gate:

Run when the diff includes source-code or code-adjacent files (.cs, .ts, .tsx, .html, .scss, .css, tests, scripts, build/config-as-code).
SKIP for docs-only / markdown-only diffs. Record: Skipped Phase 3.5 — no code files in diff.

Protocol:

Set the [Review Phase 3.5] task to in_progress.
Invoke $code-simplifier scoped to the changed code files only (pass the Phase 1 diff source — working-tree, staged, branch, or commit range — so it refines the related changed files, NOT the whole codebase). Direct it to surface reuse, DRY, KISS/YAGNI, naming, dead-code, altitude/layer-placement, and readability simplifications.
Capture, do NOT auto-apply. In review context, $code-simplifier runs in report mode: integrate its recommendations into the main report under ## Code-Simplifier Optimization Findings with file:line evidence and a one-line rationale each. These are findings, not edits.
Set the [Review Phase 3.5] task to completed.

Pipeline integration: Phase 3.5 findings are ordinary findings — they consolidate in Phase 4, are validated in Phase 6 ($why-review --validate-findings filters false-positive or change-cost-raising simplifications), and only validated ones are fixed in Phase 7. NEVER let $code-simplifier mutate the working tree before Phase 6 validates its suggestions.

Parent workflow boundary: When this skill is invoked as step 1 inside $workflow-review-changes, still run Phase 3.5 (it is a review dimension, producing findings for the report) but do NOT fix here — the parent workflow's $code-simplifier self-review and $feature-implement fix cycle own application. Record the findings and hand the report to parent step 2.

Phase 3.7: Integration-Test-Review Coverage Gate (MANDATORY when behavior-bearing code changed)

Purpose: Phases 2-3 prove the change is correct as written; this gate proves the change is covered and specced. $integration-test-review is the canonical owner of the 7-gate test-quality audit — its Gate 7 (Change Coverage) maps every behavior-changing production file in the diff to a covering test (integration-first; unit fallback needs explicit justification) AND a spec TC. This skill delegates to it rather than duplicating that logic. SYNC:integration-test-sync-check stays as the lightweight file-pairing check; this gate goes deeper — assertion quality, data-state verification, repeatability, and bidirectional spec↔test↔code alignment over the full change set.

Entry gate:

Run when the diff includes behavior-bearing source code: handlers, commands, queries, services, entities, event consumers, controllers, background jobs, or frontend logic.
SKIP for docs-only / markdown-only / pure styling-asset diffs. Record: Skipped Phase 3.7 — no behavior-bearing code in diff.

Protocol:

Set the [Review Phase 3.7] task to in_progress.
Invoke $integration-test-review scoped to the Phase 1 diff source (working-tree, staged, branch, or commit range) so it audits the FULL change set — changed production code AND changed test files, never just the test files. It runs all 7 quality gates, builds the Gate 7 Coverage Mapping Table, and cross-checks spec TCs in both directions.
Capture, do NOT auto-fix. Integrate its output into the main report under ## Integration-Test-Review Findings: per-gate verdicts, the Coverage Mapping Table, and every GAP / SPEC-GAP / unjustified COVERED-UNIT as a finding (GAP = HIGH severity minimum; CRITICAL for auth/money/data-integrity paths).
Set the [Review Phase 3.7] task to completed.

Pipeline integration: Phase 3.7 findings are ordinary findings — consolidated in Phase 4, validated in Phase 6, fixed in Phase 7. GAP fixes WRITE the missing test via $integration-test; SPEC-GAP fixes run $spec [mode=tests] [update]. The Phase 7 restart then re-audits coverage over the full updated diff, including the new tests.

Parent workflow boundary: When this skill is invoked as step 1 inside $workflow-review-changes, do NOT run Phase 3.7 locally — the parent workflow's dedicated $integration-test-review step owns the 7-gate audit and coverage mapping. Record Phase 3.7 deferred to parent workflow $integration-test-review step. (SYNC:integration-test-sync-check still applies locally as the lightweight pairing check.)

Phase 4: Generate Final Review Result

Update report with final sections (MUST ATTENTION — include every section below):

Overall Assessment (big picture summary)
Critical Issues (must fix before merge)
High Priority (should fix)
Architecture Recommendations
Documentation Staleness (list stale docs with what changed, or "No doc updates needed")
Spec Drift Adjudication (per behavior-changing file: CODE-WRONG / SPEC-STALE / AMBIGUOUS / SPEC-SILENT / Spec in sync, with the routed fix; or "No behavior change — N/A")
Dual-Feedback Ledger (REQUIRED — see below; or "No behavior change — N/A")
Positive Observations
Suggested commit message (based on changes)

Dual-Feedback Ledger (REQUIRED for every behavior-changing finding). A behavior gap must feed back into BOTH the spec AND the tests — not merely fix the code. The Spec Drift Adjudication row above and the Phase 3.7 Gate 7 coverage row each cover only ONE axis; this ledger unifies them into a single "update BOTH" assertion so neither is silently skipped. For each behavior-changing finding, emit one row with two cells:

| Finding (file:line) | Spec feedback | Test feedback | | --------------------- | ---------------------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------- | | {cite} | {the §8 spec/Feature-Spec update needed — e.g. $spec [update] then $spec [mode=tests], OR N/A-because-CODE-WRONG-spec-already-correct} | {the TC/regression test needed — e.g. new §8 regression TC via $spec [mode=tests], or covering integration test via $integration-test} |

A blank cell on EITHER axis = FAIL. "N/A" alone is not allowed — every N/A must carry its reason inline (e.g. N/A — CODE-WRONG: canonical spec already describes the correct behavior, so no spec edit; only the regression TC is owed).

CODE-WRONG finding → Spec feedback is typically N/A — spec correct, Test feedback is REQUIRED (regression TC first, per SYNC:spec-drift-adjudication).

SPEC-STALE finding → BOTH cells are non-N/A: Spec feedback = $spec [update] then $spec [mode=tests]; Test feedback = the new/updated TC + guarding test.

SPEC-SILENT finding (code correctly enforces an invariant the spec never states) → BOTH cells are non-N/A: Spec feedback = add the missing §4 BR / §3 AC (+ §5 invariant if applicable) and a §8 TC via $spec [update] + $spec [mode=tests]; Test feedback = the new property/regression test guarding the now-written invariant. The highest-value capture — never leave a discovered invariant only in code or only in tests.

Covered-but-stale TC (Gate 7 SPEC-GAP routed by $integration-test-review — its Gate 7 classifies a TC that exists but no longer describes current behavior as a SPEC-GAP, not a satisfied coverage row): Spec feedback = correct the stale §8 TC via $spec [mode=tests] [update]; Test feedback = update the guarding test to the corrected TC.

This ledger is itself an ordinary finding set: it consolidates here (Phase 4), is validated in Phase 6 ($why-review --validate-findings confirms BOTH axes are present for each behavior change), and its owed spec/test actions are applied in Phase 7. A ledger row with a blank axis that survives to Phase 7 = the review is INCOMPLETE.

Phase 5: Docs-Update Triage (CONDITIONAL)

If Documentation Staleness Check in Phase 4 identified stale docs:

Record impacted documentation and the proposed sync/update path in the review report
Add each stale-doc item to the Phase 6 findings validation payload
Do NOT invoke $docs-update yet; stale-doc findings are fixed in Phase 7 only after $why-review --validate-findings returns CLEAN for them
If Phase 7 later applies doc fixes, the next recursive $review-changes invocation must re-review the updated docs from Phase 0

Phase 5 triages only the docs the review FLAGGED. Regardless of whether anything is flagged here, the mandatory Phase 8 final $docs-update gate still runs once the review converges clean — it independently detects impacted docs this triage may have missed. Phase 5 is conditional; Phase 8 is unconditional.

Readability Checklist (MUST ATTENTION evaluate)

Before approving, verify artifacts are easy to read, maintain, understand:

Schema visibility — Function computes data structure? Comment shows output shape so readers don't trace code
Non-obvious data flows — Data transforms through multiple steps? Brief comment explains pipeline
Self-documenting signatures — Params explain their role; flag unused params
Magic values — Unexplained numbers/strings → named constants or inline rationale
Naming clarity — Variables/functions reveal intent without reading implementation

Review Checklist

1. Architecture Compliance (MUST ATTENTION)

Follows project's layer/module boundaries (read docs/project-config.json or equivalent)
No cross-module/service direct data access where boundaries exist
Logic placed in lowest responsible layer (not in orchestrators/top-layer classes)

2. Code Quality & Clean Code (MUST ATTENTION)

Single Responsibility Principle — each function/class does ONE thing
No code duplication (DRY) — grep for similar code, extract if 3+ occurrences
Appropriate error handling following project patterns
No magic numbers/strings (extract to named constants)
Type annotations on all functions (where language requires)
Early returns/guard clauses used
YAGNI — no speculative features, unused parameters, premature abstractions
KISS — simplest solution meeting requirement
Follows existing codebase conventions (verify with grep for 3+ examples)

2.5. Naming Conventions (MUST ATTENTION)

Names reveal intent (WHAT not HOW)
Specific names, not generic (orderRecords not data)
Booleans: prefix with state-indicating verb (isActive, hasPermission, canEdit)
No cryptic abbreviations

3. Project-Specific Patterns (MUST ATTENTION)

Read project's patterns/conventions reference docs BEFORE flagging violations
Verify 3+ existing examples before concluding a pattern is a violation
Flag deviation from project patterns with evidence (file:line showing existing pattern)

4. Security (MUST ATTENTION)

No hardcoded credentials, tokens, or secrets
Proper authorization checks at all entry points
Input validation at system boundaries (user input, external APIs, message payloads)
No injection risks (SQL, command, template, etc.)

5. Performance (MUST ATTENTION)

No O(n²) complexity where O(n) or O(1) is possible (use lookup structures)
No N+1 query patterns (batch load related data before iterating)
Pagination for all list queries (never fetch unbounded result sets)
Parallel operations where independent (not forced sequential)
Async/await used correctly (no blocking in async context)
Query patterns have appropriate indexes

6. Common Issues (MUST ATTENTION)

Unused imports or variables
Debug/logging statements left in that should not be in production
Hardcoded values that should be configuration
Missing async/await or promise handling
Incorrect or absent exception handling
Missing validation at boundaries

6.5 Bugfix Debugger Trace Gate (MUST ATTENTION)

For bugfix, failed-verification, stale/incorrect final output, regression, or behavior-changing fixes, FAIL review if any required proof is missing:

Debugger Trace: End -> Start names the observed final state and final reader/query/renderer/assertion
backward hops are evidenced from reader -> storage/projection/cache -> writer -> consumer/handler/job -> producer/origin
all feeder paths that can write the final state are enumerated or explicitly marked unknown
hypothesis matrix classifies root causes as primary, contributing, ruled out, latent, or unknown
owning fix layer is justified as the lowest shared owner, not the symptom site by default
forward convergence proof and regression test/proof mapping show why the final symptom cannot persist

7. Documentation Staleness (MUST ATTENTION)

For each changed file: identify related docs (feature docs, architecture references, READMEs)
Changed logic → verify relevant feature/module docs still accurate
Changed tooling (scripts, configs, CI) → verify setup/getting-started docs still accurate
New feature/component added → flag if corresponding doc missing
Test specs reflect current behavior after changes
API changes reflected in relevant API docs or specs
Spec-drift adjudication (SYNC:spec-drift-adjudication): for every behavior-changing file, decide whether a divergence from the canonical Feature Spec is CODE-WRONG (change is the defect — BLOCKING, fix code/test), SPEC-STALE (change is intended — update spec via $spec [update] first), AMBIGUOUS (intended behavior unclear — a direct user question before editing either side), or SPEC-SILENT (code correctly enforces an invariant no spec artifact states — enrich: add §4 BR/§3 AC + §8 TC + guarding test). Do not flag a divergence as a one-directional "stale doc" without naming which side is canonical. Unadjudicated behavior-vs-spec divergence is a FAIL; an unwritten-but-enforced invariant left uncaptured is equally a FAIL.

8. M1-M6 Compliance Gate — Code-to-Spec Drift (BLOCKING, MUST ATTENTION)

Contract: See .claude/skills/shared/sdd-artifact-contract.md → "AI-SDD Mandates (M1-M6)". This review enforces M6 for any spec/feature-doc/PBI/story/test-spec touched by — or supposed to be synced by — this change. Frame each check as: did this change introduce M1/M2 prose leakage, break a logical-ID mapping (M3), or create AC/expected-result ambiguity (M4)? A FAIL must name the violated mandate ID and cite the changed file + line. Passing an introduced M1-M5 violation makes this review itself defective.

Carriers are EXEMPT from M1/M2 — source identifiers stay CORRECT inside [Source: ...], **Evidence**, **IntegrationTest** fields, YAML frontmatter, and ```mermaid ``` blocks. Only flag leakage in spec/doc narrative prose. Banned prose token list: docs/project-reference/spec-principles.md §3.2. Scope this gate to changed artifact files (docs/specs/**, PBI/story/test-spec files in the diff); SKIP with a one-line note when the diff touches no such artifact.

M1 — No introduced tech leakage in prose. FAIL if the diff adds a framework/product, language-native type, or product/design-pattern class name to spec/doc narrative prose, headings, or AC text (banned list in spec-principles.md §3.2). Cite the changed file + line + token.
M2 — No introduced source code in prose. FAIL if the diff expresses a requirement as a class/method/file-path/namespace used as a noun instead of a business operation. Source identifiers belong only in evidence carriers. Cite the changed line.
M3 — Logical-ID mapping preserved. FAIL if the change adds a requirement/rule/TC without a logical ID (FR-/BR-/OP-/TC-), strips a logical ID, demotes it below the [Source:] evidence, writes physical code coordinates or repository-root paths instead of a stack-portable abstract anchor ([Source: namespace/service/id]), OR drops the [Source:] abstract-anchor evidence (evidence is REQUIRED and KEPT — SECONDARY to the logical ID; a code move alone does NOT change the anchor — physical coords live only in the provenance sidecar).
M4 — No introduced AC ambiguity. FAIL if the change leaves an AC/expected-result vague ("handle appropriately", "process normally", "as needed"), implementable two different ways while both claim conformance, or with no observable completion state / named error condition.
M5 — Spec stays rebuildable. FAIL if the change makes the spec/doc depend on reading the new code to be understood (a zero-codebase-knowledge team could no longer re-implement on a different stack from the artifact alone). Cite the file + missing detail.

If ANY item fails → the verdict is FAIL; list each violated mandate ID with its changed-file/line citation in the Critical Issues or High Priority section.

Output Format

Provide feedback in this format:

Summary: Brief overall assessment

Critical Issues: (Must fix before commit)

Issue 1: Description and suggested fix

High Priority: (Should fix)

Issue 1: Description

Suggestions: (Nice to have)

Suggestion 1

Documentation Staleness: (Docs that may need updating)

Doc 1: What is stale and why
No doc updates needed — if no changed file maps to a doc

Spec Drift Adjudication: (Behavior-changing changes only — per SYNC:spec-drift-adjudication)

<behavior/file> → CODE-WRONG | SPEC-STALE | AMBIGUOUS | SPEC-SILENT — verdict + routed fix ($spec [update], regression TC, a direct user question, or enrich-spec: add §4 BR/§3 AC + §8 TC + guarding test)
Spec in sync — if changed behavior matches the canonical Feature Spec
No behavior change — N/A — if the diff is docs/tooling/style only

Debugger Trace Gaps: (Bugfix/behavior-changing changes only)

Trace complete — if the required trace, feeder paths, hypothesis matrix, owner, and forward proof are present
Gap 1: Missing or weak trace evidence and why it blocks PASS

Goal Satisfaction: (MANDATORY before any PASS verdict — resolve the active Goal Contract per the goal-contract-satisfaction-loop protocol: active plan goal.md → plans/goals/{YYMMDD-HHmm}-{slug}/goal.md; if none exists, record No active goal — skipped: {one-line reason})

| Success Criterion | Evidence | Status | | ----------------- | ---------------------------------------- | ----------------- | | {saved criterion} | {file:line, command output, report path} | PASS/FAIL/BLOCKED |

Overall PASS is BLOCKED while any required criterion is FAIL — a code-quality-clean review that misses the saved goal is NOT a PASS.
BLOCKED status requires a user-facing escalation reason recorded in the matrix row and the goal file.
Cite evidence references; never restate the goal text or copy secrets/sensitive payloads into the matrix or goal file.
After the verdict, update the goal file: append an Iteration Log entry and sync its Goal Satisfaction matrix.

Positive Notes:

What was done well

Suggested Commit Message:

type(scope): description

- Detail 1
- Detail 2

Systematic Review Protocol (for 10+ changed files)

When Phase 1 finds 10+ changed files, apply the Systematic Review Batching protocol (map-reduce: size-capped batches + hierarchical synthesis) defined below.

Workflow Recommendation

MANDATORY — NO EXCEPTIONS: If NOT already in a workflow, MUST use a direct user question to ask user. Do NOT judge task complexity or decide "simple enough to skip" — user decides, not you:

Activate workflow-review-changes workflow (Recommended) — run the canonical workflow from .claude/workflows.json; it sequences this skill, findings validation, parallel reviewers, code-simplifier self-review, fix-plan cycle, full re-review restart, docs, and handoff.

Execute $review-changes directly — run this skill standalone

Architecture Boundary Check

For each changed file, verify no import from forbidden layer:

Read rules from docs/project-config.json → architectureRules.layerBoundaries
Determine layer — For each changed file, match path against each rule's paths glob patterns
Scan imports — Grep file for import statements
Check violations — If any import path contains layer name listed in cannotImportFrom, it is a violation
Exclude framework — Skip files matching any pattern in architectureRules.excludePatterns
BLOCK on violation — Report as critical: "BLOCKED: {layer} layer file {filePath} imports from {forbiddenLayer} layer ({importStatement})"

If architectureRules not present in project-config.json, skip silently.

Phase 6: Why-Review Findings Validation Gate (MANDATORY before fixing findings)

Purpose: Validate own findings BEFORE any fix. Verify EVERY finding is correct, proof-backed (file:line), reasonable, and convention-aligned. Catch false positives, inflated severity, and missed improvements before code/doc edits.

MANDATORY: REQUIRED todo task whenever findings exist. Register via task tracking as [Review Phase 6] Why-review findings validation gate (already in Phase task list above). Do NOT fix, docs-update, commit, or hand off until this gate passes CLEAN or reaches an explicit blocked state.

Trigger: Any finding produced (Critical, High, Medium, OR Low). Skip ONLY when report verdict is unconditional PASS with literally zero findings.

UNCONDITIONAL INVOCATION: If even one finding exists, the $why-review skill MUST be invoked via the skill invocation before any fix, docs-update, commit, or handoff. There is NO inline alternative — manually re-reading the cited file:lines, re-tracing in your head, or declaring the findings "already validated" does NOT count. The only way to pass this gate is an actual $why-review --validate-findings skill call that returns a verdict.

Parent workflow boundary: When this skill is invoked as step 1 inside $workflow-review-changes, do NOT run this Phase 6 locally. Stop after the review report and hand it to parent workflow step 2; the parent runs $why-review --validate-findings before any parallel reviewers or fixes.

Protocol (capped re-do loop):

Read own finalized report from plans/reports/{skill}-{date}-{slug}.md
Invoke the $why-review skill via the skill invocation (terminal --validate-findings mode — runs in the SAME main-agent session, never spawns a sub-agent, never recurses). "Same session" means the why-review skill executes in this conversation — it does NOT mean you may substitute your own inline re-reading for the skill call. The skill MUST actually run. Pass arg: --validate-findings plans/reports/{skill}-{date}-{slug}.md — for EACH finding verify (a) file:line proof exists and is accurate, (b) the finding is correct (re-trace the cited code), (c) severity is reasonable and not inflated, (d) it reflects project best practices/conventions; steel-man each rejected interpretation; and surface any MISSED finding or enhancement opportunity the review overlooked
Read the validation verdict path returned by why-review, expected as plans/reports/why-review-validate-{date}.md
Classify the why-review verdict:
- CLEAN — all findings confirmed correct / proof-backed / reasonable / best-practice, AND no new finding issue or enhancement opportunity surfaced → append ## Why-Review Validation line to own report ("All N findings re-validated against actual code; no changes."), gate PASSES; if N > 0, proceed immediately to Phase 7.
- HAS ISSUES — why-review demotes/removes a finding, flags a missing or inaccurate proof, OR surfaces a new finding issue / enhancement opportunity → go to step 5.
Reconcile: UPDATE own finalized report — revise severities, remove false positives, add the surfaced findings/enhancements, and record a ## Why-Review Validation Notes section citing what changed and why.
RE-DO $why-review --validate-findings on the UPDATED report (return to step 2) — re-validation is required ONLY because the report changed. Each pass is terminal (validate mode never recurses); the loop is owned and bounded HERE. Repeat until a why-review round comes back CLEAN, or max 2 re-do rounds (3 total validate passes) is reached.
If still not CLEAN after the cap: record unresolved items under ## Why-Review Validation — Unresolved and escalate to the user via a direct user question instead of silently looping.

Skip conditions (record explicit reason if skipping):

Verdict is unconditional PASS with zero findings → log "Skipped — no findings to validate" and do NOT run Phase 7
$why-review itself is the active skill context → do NOT recurse; why-review re-validates via its own terminal --validate-findings mode (see its Findings Validation Gate)

Why this exists: AI reports can inherit confirmation bias, false positives, and severity inflation. Validation proves findings before edits; re-validation after report changes closes the gap where corrected findings are never checked again.

Phase 7: Recursive Auto-Fix + Full Re-Review Loop (MANDATORY when validated findings remain)

Purpose: Fixes change the review target. Next check MUST be a full new $review-changes invocation from Phase 0, not continuation from old review state.

Trigger: Phase 6 returns CLEAN and the validated report still contains one or more findings, weaknesses, stale-doc items, missing-test items, or required improvements.

Parent workflow boundary: When this skill is invoked as step 1 inside $workflow-review-changes, do NOT auto-fix or re-invoke $review-changes from here. Parent workflow steps 10-15 own $plan, $plan-review, $plan-validate, $why-review, $feature-implement, and the full restart gate.

Protocol:

Create fresh fix-cycle tasks before editing: one task per validated finding, one targeted-verification task, one $review-changes restart task.
Auto-fix validated findings at the owning layer. Stale docs: run $docs-update or edit canonical docs only after validation. Tests/specs: update canonical artifact before derived dashboards.
Run targeted verification for the fix set: tests, lint, docs/spec sync, SDD, graph, or config checks as applicable.
Append ## Fix Cycle {N} to the review report: findings fixed, files changed, verification commands/results, and unresolved items with reasons.
Re-invoke $review-changes in the SAME main-agent session on the full current review target:
- Create brand-new task list for all phases
- Re-run Phase 0 blast radius, Phase 0.3 risk detection, Phase 0.7 surface categorization, Phase 1 diff collection, and later phases
- Re-read all changed files from scratch, including original changes and Phase 7 fixes
- Treat previous report as historical context only; never reuse prior findings as truth
Repeat Phase 0 → Phase 7 until one complete $review-changes invocation produces unconditional PASS with zero findings and Phase 6 is skipped as "no findings to validate".

Stop conditions:

If the same validated finding repeats for 3 full review invocations with no observable progress, stop and ask the user for a decision instead of spinning.
If a finding cannot be safely auto-fixed without product/owner input, record the blocker and ask the user.
If required verification tools or sub-skills are unavailable, stop and ask before adapting the protocol.

Non-negotiable rules:

NEVER fix findings before $why-review --validate-findings confirms the current finding set.
NEVER mark review clean after a fix without rerunning the full $review-changes protocol from Phase 0.
NEVER review only the fixed files after a fix; review the full current diff because fixes can interact with earlier changes.
NEVER reuse old todo tasks after restart; each recursive review invocation breaks down all phases again.
NEVER declare unconditional PASS without the Output Format's Goal Satisfaction matrix showing every required saved criterion PASS (or BLOCKED with a user-facing escalation reason). A required-criterion FAIL is a validated finding for this fix loop.

Phase 8: Mandatory Final Docs-Update Gate (MANDATORY — always runs after the review/fix loop converges clean)

Purpose: Guarantee no stale docs survive the change. Phases 5-7 fix only docs the review flagged as findings; this terminal gate runs $docs-update unconditionally so impacted docs the dimensional review never surfaced still get reconciled against the actual changes. A clean code-review verdict does NOT imply docs are current.

Trigger: The review has converged — one full $review-changes pass produced zero findings and all validated fixes are applied. This gate ALWAYS runs in standalone mode; it is NOT gated on a flagged staleness finding.

Parent workflow boundary: When this skill is invoked as step 1 inside $workflow-review-changes, do NOT run Phase 8 locally — the parent workflow's own $docs-update step (after $feature-implement and the restart gate) owns the final docs sync. Record Phase 8 deferred to parent workflow $docs-update step.

Protocol:

Set the [Review Phase 8] task to in_progress.
Invoke $docs-update over the FULL changeset (the Phase 1 diff source plus any Phase 7 fixes). Let it detect impacted docs from the changes — feature docs, architecture references, READMEs, API docs, test specs, setup/getting-started guides.
If the Phase 4 Spec Drift Adjudication (SYNC:spec-drift-adjudication) returned any SPEC-STALE verdict — the canonical Feature Spec no longer reflects the intended behavior (includes the post-bugfix case where the spec documents the bug as correct behavior) — run $spec [update] BEFORE $docs-update so the spec is corrected to intended behavior first. Never let $docs-update codify broken or superseded behavior. CODE-WRONG verdicts are NOT a spec edit — they were already fixed in the Phase 7 code-fix loop. Any SPEC-SILENT verdict — an invariant the code already enforces but no spec artifact states — is an ENRICHMENT: run $spec [update] (add the §4 BR/§3 AC) + $spec [mode=tests] (add the §8 TC) BEFORE $docs-update, then ensure a guarding test exists; never leave the discovered invariant unwritten.
Record applied doc updates (or No impacted docs — verified N changed files against related docs) under ## Phase 8 Docs-Update in the review report.
Set the [Review Phase 8] task to completed.

Termination guarantee: Distinguish two edit kinds in Phase 8. Docs PROSE edits (narrative/reference doc text, no new spec rule) do NOT re-trigger the loop (no code behavior changed); they ARE subject to the M1-M6 spec-drift check (Review Checklist §8) and a final read-back — termination preserved. SPEC-CONTENT edits — a newly WRITTEN spec rule (a SPEC-SILENT invariant promoted to §3/§4/§8, or a SPEC-STALE correction that changes documented intent) — trigger exactly ONE bounded, module-scoped re-review of the whole package (spec + tests + code for the affected module) against the enriched spec, to confirm the newly-written rule is actually enforced in code and guarded by a test, and to surface any further hidden rule. That single bounded pass terminates unless it itself produces a new validated finding (which then enters the normal loop). Termination stays guaranteed: the bounded pass is module-scoped and runs at most once per enrichment — not a full Phase-0 restart — and a clean bounded pass ends the skill. This converts enrichment from "fires once at the end, never rechecked" into "fires, then gets one convergence pass," instead of looping review↔docs forever.

MANDATORY: Never declare the review complete or hand off until Phase 8 has run (or been explicitly deferred to the parent workflow). A passing review with skipped docs-update is an INCOMPLETE review.

Next Steps

MANDATORY — NO EXCEPTIONS after completing this skill, MUST use a direct user question to present options. Do NOT skip because task seems "simple" or "obvious" — user decides:

"$code-review (Recommended)" — Deeper code quality review
"$watzup" — Wrap up session and review all changes
"Skip, continue manually" — user decides

AI Agent Integrity Gate (NON-NEGOTIABLE)

Completion ≠ Correctness. Before reporting ANY work done, prove it:

Grep every removed name. Extraction/rename/delete touched N files? Grep confirms 0 dangling refs across ALL file types.

Ask WHY before changing. Existing values are intentional until proven otherwise. No "fix" without traced rationale.

Verify ALL outputs. One build passing ≠ all builds passing. Check every affected stack.

Evaluate pattern fit. Copying nearby code? Verify preconditions match — same scope, lifetime, base class, constraints.

New artifact = wired artifact. Created something? Prove it's registered, imported, reachable by all consumers.

Related Skills

| Skill | Relationship | When to Call | | ----------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------ | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | $docs-update | Mandatory terminal gate (Phase 8) — final docs sync after the review/fix loop converges; also the primary fix path for flagged staleness | ALWAYS at Phase 8 once review is clean (standalone) — unconditional; AND during Phase 7 for validated staleness findings. Deferred to parent in $workflow-review-changes. | | $spec-index | Derived index — regenerates the bucket INDEX.md/ERD FROM the Feature Specs (never a source of truth) | After specs change, to refresh navigation aids — NOT for correcting specs | | $spec [update] | Canonical spec updater — corrects feature doc §1-8 (the single source of truth) | Called internally by docs-update; call directly for targeted update — and BEFORE docs-update if a spec-was-wrong scenario is detected | | $spec [mode=tests] [update] | Test spec updater — called when test cases may be stale | Called internally by docs-update; call directly for targeted test case update | | $integration-test-review | Mandatory coverage gate (Phase 3.7) — 7-gate test-quality audit + Gate 7 change-coverage mapping (every behavior change → covering test + spec TC) | ALWAYS at Phase 3.7 when behavior-bearing code changed (standalone); deferred to the parent's dedicated step in $workflow-review-changes. Skip only docs-only diffs | | $review-ui | UI/frontend quality gate — overflow, responsive flex, z-index, SCSS/BEM | Owned by this skill — invoked internally as the UI dimension (ui-ux-designer sub-agent) when the diff has frontend/UI files; NOT a separate workflow step | | $code-simplifier | Quality-optimization dimension — clarity/consistency/maintainability simplifications | Owned by this skill — invoked internally in Phase 3.5 (report mode) when the diff has code files; its findings flow through Phase 6 validation → Phase 7 fix | | $code-review | Code quality — deeper review of changed code | Always follows review-changes quality pass |

Standalone Chain

When called outside a workflow (i.e., user ran $review-changes directly):

review-changes (you are here)
  │
  ├─ Phase 3.5: Code-simplifier optimization (INTERNAL — $code-simplifier over changed code files, report mode)
  │    → Simplification findings feed Phase 6 validation → Phase 7 fix (skip docs-only diffs)
  │
  ├─ Phase 3.7: Integration-test-review coverage gate (INTERNAL — $integration-test-review over the FULL diff, 7 gates)
  │    → Gate 7 maps every behavior-changing production file to a covering test (integration-first) + spec TC
  │    → GAP/SPEC-GAP verdicts feed Phase 6 validation → Phase 7 fix
  │    → GAP fix = WRITE the missing test via $integration-test; SPEC-GAP fix = $spec [mode=tests] [update] (skip docs-only diffs)
  │
  ├─ Follow-on quality checks (review-architecture → code-review → performance)
  │
  ├─ Phase 5: Documentation Staleness Triage
  │    → If stale docs detected: [REQUIRED] include as finding for Phase 6 validation
  │    → If validated in Phase 6: [REQUIRED] fix in Phase 7 via $docs-update or canonical doc edit
  │    → Then recursively restart $review-changes from Phase 0
  │
  ├─ Integration test check (SYNC:integration-test-sync-check):
  │    → If logic changes touch tested areas: [REQUIRED] → $integration-test [from-changes]
  │    → Then: $integration-test-review → $integration-test-verify
  │
  ├─ Translation sync check (SYNC:translation-sync-check):
  │    → If multilingual UI text changes lack locale updates: [REQUIRED] ask the user directly + explicit decision
  │
  ├─ Spec drift adjudication (SYNC:spec-drift-adjudication) — ALL behavior-changing reviews:
  │    For every behavior-bearing change diverging from the canonical Feature Spec, classify:
  │    → CODE-WRONG (change violates an intended spec rule/AC/invariant) → [REQUIRED] BLOCKING finding; fix code/test (regression TC first)
  │    → SPEC-STALE (intentional new behavior the spec no longer reflects) → [REQUIRED] $spec [update] BEFORE $docs-update, then $spec [mode=tests] [update]
  │    → AMBIGUOUS → [REQUIRED] ask the user directly (or canonical spec owner) before editing either side
  │    → SPEC-SILENT (code enforces a correct invariant the spec never states) → [REQUIRED] enrich: $spec [update] add §4 BR/§3 AC + $spec [mode=tests] add §8 TC, then guarding test
  │    Bugfix sub-case: if post-bugfix AND spec documents the bug as expected behavior → SPEC-STALE; never let $docs-update codify broken behavior.
  │    Never normalize drift just because code/tests are green.
  │
  ├─ Phase 6 + Phase 7 recursive loop
  │    → If ANY findings exist: $why-review --validate-findings
  │    → If validated findings remain: auto-fix, verify, then restart $review-changes from Phase 0
  │    → Repeat until a full review invocation has zero findings
  │
  ├─ Phase 8: [MANDATORY FINAL after zero findings] → $docs-update over the full changeset
  │    → ALWAYS runs (unconditional) — syncs every impacted doc so none stay stale
  │    → docs PROSE edits (no new spec rule); M1-M6 check + read-back, NO full code re-review (guarantees termination)
  │    → SPEC-CONTENT edits (a newly written spec rule — SPEC-SILENT promoted to §3/§4/§8, or a SPEC-STALE correction changing documented intent) → exactly ONE bounded, module-scoped re-review of the affected module's package (spec + tests + code) to confirm the new rule is enforced in code + guarded by a test; that single pass terminates unless it itself yields a new validated finding (then normal loop). At most once per enrichment, never a full Phase-0 restart.
  │    → A passing review with skipped docs-update is an INCOMPLETE review
  │
  └─ [RECOMMENDED after Phase 8] → $watzup
        Summary of all review findings, doc changes, and test coverage status.

[CRITICAL — TOP 3 RULES]

MUST ATTENTION Phase 0 graph blast-radius FIRST — NEVER skip; informs entire review order

Findings trigger validate → fix → full restart. Run $why-review --validate-findings, fix validated findings, then rerun $review-changes from Phase 0 until a full pass has zero findings.

MUST ATTENTION task tracking ALL phases before starting; missing tests MUST surface via a direct user question — NOT silently logged

[IMPORTANT] Use task tracking to break ALL work into small tasks BEFORE starting — including tasks for each file read. Prevents context loss from long files. For simple tasks, AI MUST ATTENTION ask user whether to skip.

Systematic Review Batching (map-reduce) — When a changeset is large, do NOT review files one-by-one. Partition into size-capped batches, fire one specialized sub-agent per batch in parallel, then reduce. This bounds EVERY context — each batch agent AND the orchestrator — so coverage stays complete as file count grows.

Trigger ladder (one ordered escalation — not competing thresholds):

< 10 changed files → sequential per-file review (default; no batching).

≥ 10 changed files → switch to systematic parallel mode. Announce: "Detected {N} changed files. Switching to systematic parallel review protocol." Then: categorize → size-capped batches → flat consolidation.

categories > 6 OR files > 40 → additionally insert the hierarchical synthesis tier (below). Everything from rung 2 still applies.

Step 1 — Categorize. Group changed files into logical categories derived from the project's actual structure (not forced). Category is the concern axis; orient with these examples, derive what fits the repository:

| Category Type | Example Groupings | | ------------------- | --------------------------------------------------------------------- | | Agent/Tooling | AI scripts, hooks, skill definitions, workflow configs, linting rules | | Root config/docs | Root README, project config, CI/CD pipeline configs | | Reference docs | Architecture docs, patterns references, setup guides | | Feature/domain docs | Business feature documentation, spec files, ADRs | | Backend logic | Service/handler/controller source (infer from project structure) | | Frontend logic | UI component/state/API source (infer from project structure) | | Data/Schema | Migrations, schema files, seed data | | Tests | Unit, integration, E2E test files | | Infrastructure | Docker, k8s, CI/CD, cloud manifests |

Step 2 — Size-capped batches. One sub-agent per batch of ≤8 files OR ≤2000 diff-lines, whichever hits first. Category stays the concern axis, but any category exceeding a cap splits into multiple size-capped batches (30 backend files → 4 batches). Size caps — not category caps — make "many files" safe: a category cap alone lets one giant category blow a single agent's context.

Step 2a — Sub-agent type per batch (match the batch's dominant concern):

Code logic (any stack) → code-reviewer

Security-sensitive changes → security-auditor

Performance-critical paths → performance-optimizer

Docs, plans, specs, configs, infra → general-purpose

Each batch sub-agent receives: its full file list; SYNC:category-review-thinking as its primary thinking model — derive each category's concerns from first principles, NOT a fixed checklist (if the consuming skill does not carry that block, apply category-first thinking directly); project reference docs relevant to its concern (discover via *patterns*, *conventions*, *style-guide*); cross-reference verification instructions (counts, tables, links). All batch agents run in parallel and write findings to plans/reports/ (per SYNC:task-tracking-external-report); reducers read from disk, never from memory.

Step 3 — Reduce.

Flat reduction (rung 2, ≤6 categories AND ≤40 files): the orchestrator collects each batch report, cross-references counts/tables/contracts ACROSS batches, detects gaps visible only across categories (feature in code but missing from docs; new API endpoint with no client call), and consolidates into one categorized holistic report.

Hierarchical reduction (rung 3, > 6 categories OR > 40 files): insert a mid-tier — each concern gets ONE synthesizer agent that reads only its own batch reports and emits a single concern-synthesis. The orchestrator reads the concern-syntheses (~5), never the raw batch reports — keeping the reducer's context O(#concerns), not O(#files).

Cross-concern interaction pass (mandatory at rung 3 — closes the synthesis-tier blind spot): concern-siloed synthesis can drop an interaction spanning two concerns AND two batches (tainted source in data-layer/batch 7 → sink in api/batch 3). So: (a) each concern-synthesizer MUST emit an explicit "cross-concern interaction candidates" list — entities/symbols/contracts it touched that plausibly bind to another concern (shared DTOs, event names, table/collection names, exported symbols); (b) the orchestrator MUST run the Step-3 cross-reference/gap step over those candidate lists across all concern-syntheses, not only within a batch, before concluding. Without this pass the tier trades completeness for context-bounding on exactly the large diffs it targets.

Step 4 — Holistic assessment. With all findings combined, judge: overall coherence as a unified intent; cross-category sync (docs match code? contracts match callers?); risk areas where categories interact; missing doc/spec updates for changed artifacts.

No silent truncation. If any cap forces sampling or a batch is dropped for budget, ANNOUNCE the dropped/sampled scope explicitly — bounded coverage must never read as complete coverage.

End-to-Start Debugger Trace — For non-trivial bugs, failed verification, regression fixes, behavior-changing code, or unclear code flow, start from the observed final state and walk backward before proposing a fix.

Frame 0: observed end state — Name the exact user-visible output, failing assertion, log line, persisted value, API response, rendered UI, or aggregate bucket. Record the reader/query/renderer that produced it with file:line evidence.

Walk backward one hop at a time — Trace final reader -> projection/cache/storage -> writer -> consumer/handler/job -> producer/caller -> original trigger. At every hop record: input, transformation, output, owner, and evidence.

Enumerate all feeder paths — Find every upstream producer/caller/event/job that can write into the final path, including retry, async, cache, background, and alternate UI/API paths. Mark each path verified, ruled out, or still unknown.

Build the hypothesis matrix — For each plausible cause, list evidence for, evidence against, how to reproduce/verify, blast radius, and status (primary, contributing, ruled out, latent). Do not fix until competing causes are explicitly resolved or bounded.

Choose the owning fix layer — Identify the invariant owner and the lowest shared point that protects all downstream consumers. A fix at the symptom site is rejected unless the symptom site owns the invariant.

Prove convergence forward — After choosing the fix, walk start -> end again and show how the corrected state reaches the observed final output. Map each root cause to a fix part and each fix part to a test/proof.

BLOCKED until: final state named · backward trace written · all feeder paths enumerated · hypothesis matrix completed · owning fix layer justified · forward convergence proof mapped to tests.

NEVER: Start at the first suspicious code path. Collapse multiple producers into one "flow". Treat duplicate symptoms as duplicate records without proving the read model. Skip ruled-out hypotheses.

Critical Thinking Mindset — Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence >80% to act. Anti-hallucination: Never present guess as fact — cite sources for every claim, admit uncertainty freely, self-check output for errors, cross-reference independently, stay skeptical of own confidence — certainty without evidence root of all hallucination.

Sequential Thinking Protocol — Structured multi-step reasoning for complex/ambiguous work. Use when planning, reviewing, debugging, or refining ideas where one-shot reasoning is unsafe.

Trigger when: complex problem decomposition · adaptive plans needing revision · analysis with course correction · unclear/emerging scope · multi-step solutions · hypothesis-driven debugging · cross-cutting trade-off evaluation.

Format (explicit mode — visible thought trail):

Thought N/M: [aspect] — one aspect per thought, state assumptions/uncertainty

Thought N/M [REVISION of Thought K]: ... — when prior reasoning invalidated; state Original / Why revised / Impact

Thought N/M [BRANCH A from Thought K]: ... — explore alternative; converge with decision rationale

Thought N/M [HYPOTHESIS]: ... then [VERIFICATION]: ... — test before acting

Thought N/N [FINAL] — only when verified, all critical aspects addressed, confidence >80%

Mandatory closers: Confidence % stated · Assumptions listed · Open questions surfaced · Next action concrete.

Stop conditions: confidence <80% on any critical decision → escalate via ask the user directly · ≥3 revisions on same thought → re-frame the problem · branch count >3 → split into sub-task.

Implicit mode: apply methodology internally without visible markers when adding markers would clutter the response (routine work where reasoning aids accuracy).

Deep-dive: see $sequential-thinking skill (.claude/skills/sequential-thinking/SKILL.md) for worked examples (API design, debugging, architecture), advanced techniques (spiral refinement, hypothesis testing, convergence), and meta-strategies (uncertainty handling, revision cascades).

Understand Code First — HARD-GATE: Do NOT write, plan, or fix until you READ existing code.

Search 3+ similar patterns (grep/glob) — cite file:line evidence

Read existing files in target area — understand structure, base classes, conventions

Run python .claude/scripts/code_graph trace <file> --direction both --json when .code-graph/graph.db exists

Map dependencies via connections or callers_of — know what depends on your target

Write investigation to .ai/workspace/analysis/ for non-trivial tasks (3+ files)

Re-read analysis file before implementing — never work from memory alone. — why: long context drifts from the file; the file is ground truth

NEVER invent new patterns when existing ones work — match exactly or document deviation. — why: divergent patterns fragment the codebase and slow every future reader

BLOCKED until: - [ ] Read target files - [ ] Grep 3+ patterns - [ ] Graph trace (if graph.db exists) - [ ] Assumptions verified with evidence

Design Patterns Quality — Priority checks for every code change:

DRY via OOP: Identify classes/modules with the same purpose, naming pattern, or lifecycle. Apply your knowledge of the project's language/framework to determine the idiomatic abstraction (base class, mixin, trait, protocol, decorator). 3+ similar patterns → extract to shared abstraction.

Right Responsibility: Logic in LOWEST layer (Entity > Domain Service > Application Service > Controller). Never business logic in controllers.

SOLID: Single responsibility (one reason to change). Open-closed (extend, don't modify). Liskov (subtypes substitutable). Interface segregation (small interfaces). Dependency inversion (depend on abstractions).

After extraction/move/rename: Grep ENTIRE scope for dangling references. Zero tolerance.

YAGNI gate: NEVER recommend patterns unless 3+ occurrences exist. Don't extract for hypothetical future use.

Anti-patterns to flag: God Object, Copy-Paste inheritance, Circular Dependency, Leaky Abstraction.

Serial Attention for Design Quality — Scan one quality dimension at a time (serial passes), not all concerns at once. — why: split attention misses violations that single-focus passes catch.

Identify applicable dimensions — Based on the code's language, domain, and patterns, determine which quality dimensions apply: DRY, SOLID principles (SRP/OCP/LSP/ISP/DIP), OOP idioms, cohesion/coupling, GRASP, Law of Demeter, CQRS invariants, etc. Your list is NOT fixed — derive from what the code actually does.

One focused pass per dimension — Dedicate single-focus attention to EACH dimension in sequence. Do NOT mix concerns across passes.

Threshold: 3+ similar patterns = MANDATORY extraction — Not optional suggestion. Flag as mandatory structural fix requiring action.

2+ violations of same kind = structural finding — Report as "pattern problem" needing architectural resolution, not a list of individual instances.

Complexity Prevention (Ousterhout) — MANDATORY. Measure code by cost of change: one business change should map to one code change. Flag ALL of the following in review:

Change amplification — small business change forces edits in >3 places → structural flaw. Count edit sites for a plausible future change (add variant, add field, add authorization). >3 = reject.

Cognitive load — reader must hold too much context to safely modify. Flag deep inheritance, long parameter lists, boolean traps, implicit ordering dependencies.

Cross-cutting duplication at entry points — logging, error handling, validation, auth, transactions reimplemented per controller/handler/route. Lift to middleware / interceptor / filter / decorator / aspect.

Leaked implementation technology — repos returning IQueryable/QuerySet/Criteria/raw cursors/ORM entities to callers. Return finished results + intent-revealing methods (GetActiveVipUsers() not Query()).

Type-switch scattering — switch/if-chains on enum/discriminator in >1 place. New variant = new file, not N edits. One factory/registry switch at the boundary OK; scattered switches = reject.

Anemic models — domain objects with only getters/setters, logic floats in services. Move invariants/behavior onto the object (order.Checkout(), not order.Status = ...).

Primitive obsession — raw string/int/decimal for account numbers, emails, money, percentages, date ranges, with re-validation at every entry. Wrap in value objects / records / structs that validate once at construction.

Inline cross-cutting concerns — authorization/tenant isolation/audit/sanitization hand-written at top of every handler. Flag intent with declarative markers (@RequirePermission("Order.Delete")), enforce once centrally.

Shallow modules — tiny class, big interface (many public methods, many flags, many ctor params) wrapping little logic. A module is deep when a small interface hides a lot of implementation. If interface ≈ implementation cost to learn → inline.

Missing base class for repeated component/handler lifecycle — 3+ forms/CRUD handlers/list views reimplementing loading/dirty/submit/pagination → extract to base class / hook / composable / mixin / trait.

Premature vs delayed abstraction — rule-of-three. First occurrence: write it. Second: notice duplication. Third: extract. Don't build generic frameworks before real variation; don't copy-paste for the 4th time.

Embedded utility logic not extracted to helpers — inline paging loops (while (hasMore) { skip += take; ... }), ad-hoc datetime math, string parsing/formatting, collection partitioning, retry/backoff loops, URL/query-string building. If the algorithm is non-trivial AND stack-generic (not business-specific), extract to util/helper/extensions and let consumers call one line. Inline duplicates → duplicated bug surface.

Logic in wrong (higher) layer — downshift to callee — business/derivation logic written in the caller when the callee owns the data. Defaults: Controller code that should be App Service. App Service code that should be Domain Service or Entity. Component code that should be ViewModel/Store/Service. Caller reaching into callee's data shape to compute something → move the computation behind an intent-revealing method on the callee. Lowest responsible layer wins (Entity > Domain Service > App Service > Controller · Model/VM > Store > Component). Higher-layer placement = duplicated logic when a sibling caller needs the same thing.

Owner owns the rule — extract on first write — if a caller inlines logic that derives, normalizes, validates, or computes from another type's data, MOVE it to the owning type. Single use is sufficient — the trigger is wrong responsibility, not duplication. Sibling callers always arrive; inline copies drift silently with no compile error and no name to grep. Common offenders: Backend — inlined rules in application-layer handlers / commands / queries / services / controllers that belong on the domain entity / value object / domain service. Frontend — inlined derivations / formatting / validation in components that belong on the model / store / view-model / API service. Fix: name the rule once as a method (static or instance) on the owning type; callers invoke by name. Future variant → SECOND named method on the owner, never an inline near-duplicate. Right responsibility first; reuse is the consequence.

Extraction target — where the named rule lives:

| Shape of the rule | Goes to | | --------------------------------------------- | ----------------------------- | | Pure function over an entity's own data | static method on the entity | | Behavior that mutates / guards entity state | instance method on the entity | | Always-true invariant on a primitive value | value object constructor | | Needs DI (repo / settings / clock) | helper class registered in DI | | Domain-agnostic algorithm reused across types | util / extension method | | Pure shape / projection conversion | DTO mapping |

Pre-commit edit-site test (reject if answer is "many"):

| Change Scenario | Should touch | | ----------------------------------------------- | ------------------------- | | Add new variant (customer type, payment method) | 1 new file | | Change HTTP error response format | 1 middleware/filter | | Add timestamp field to every persisted entity | 1 base entity/interceptor | | Add authorization to a new endpoint | 1 declarative marker | | Swap database/ORM | Data layer only | | Change business calculation rule | 1 method on owning entity | | Add loading indicator pattern to forms | 1 base component/hook | | Add validation rule to a domain primitive | 1 value-object ctor | | Change paging/retry/datetime algorithm | 1 helper/util function | | Change a derivation of entity data | 1 method on the entity |

Operating heuristics:

Write the call site first.

Count edit sites for plausible future change.

Prefer removing code over adding it.

Surface assumptions at boundaries, hide details inside.

Pre-reuse scan — before writing a non-trivial block, grep for similar algorithms (while.*skip, DateTime.*Add, split/join chains, paging loops, retry loops). Match existing helper → call it. None exists but pattern is stack-generic → extract to util before second caller appears.

Layer placement test — ask "if a sibling caller needed this tomorrow, would they re-derive it?" If yes, the logic is in the wrong layer. Move it down.

Open-case-for-future-reuse — if reviewer spots a block that is likely to appear in another feature (domain-agnostic algorithm, shared lifecycle, recurring derivation), do NOT rationalize with pure YAGNI. Either extract now (if cheap) or create a tracked TODO with the exact extraction target so the second caller does not duplicate silently. Silent duplication is the default failure mode.

When in doubt ask: "What would need to change if the requirement shifts?"

The measure of good code is the cost of change. Not shortest. Not cleverest. Not most abstracted. Cheapest to safely modify having read a small local portion.

Validated-Finding Fix + Full Re-Review Loop — Re-review is triggered by a validated finding fix cycle, not by a round number. Review purpose: review → validate findings → fix validated findings → full re-review until a complete review pass finds no issues. A clean review ENDS the loop — no further rounds required.

Round 1: Main-session review. Read target files, build understanding, note issues. Output findings + verdict (PASS / FAIL).

Decision after Round 1:

No issues found (PASS, zero findings) → review ENDS. Do NOT spawn a fresh sub-agent for confirmation.

Issues found (FAIL, or any non-zero findings) → run the active review skill's findings-validation gate first; for review skills the default gate is $why-review --validate-findings <report-path>. Fix only validated findings, then restart the full review protocol from the beginning with a fresh task breakdown.

Fresh full re-review after every fix cycle: Re-run the whole review protocol over the current full target. When sub-agents are part of that protocol, spawn NEW spawn_agent calls — never reuse prior agents. Reviewers re-read ALL files from scratch with ZERO memory of prior rounds. See SYNC:fresh-context-review for the spawn mechanism and SYNC:review-protocol-injection for the canonical Agent prompt template. Each fresh full review must catch:

Cross-cutting concerns missed in the prior round

Interaction bugs between changed files

Convention drift (new code vs existing patterns)

Missing pieces that should exist but don't

Subtle edge cases the prior round rationalized away

Regressions introduced by the fixes themselves

Loop termination: After each full re-review, repeat the same decision: clean → END; issues → validate findings → fix → restart from the first review phase. Continue until a complete review pass finds zero issues. If the same validated finding repeats for 3 full invocations with no progress, or a fix requires product/owner input, escalate via a direct user question.

Rules:

A clean Round 1 ENDS the review — no mandatory Round 2

NEVER fix unvalidated findings; validate first using the caller's validation gate

NEVER skip the full re-review after a fix cycle (every fix invalidates the prior verdict)

NEVER reuse a sub-agent across rounds — every iteration that uses sub-agents spawns NEW Agent calls

Main agent READS sub-agent reports but MUST NOT filter, reinterpret, or override findings

No arbitrary sub-agent-round cap replaces the clean-review requirement; use the 3 repeated-no-progress blocker rule only to avoid infinite spinning

Track recursive invocation count and repeated blockers in conversation context (session-scoped)

Final verdict must incorporate ALL rounds executed

Report must include ## Round N Findings (Fresh Sub-Agent) for every round N≥2 that was executed.

Fresh Context Re-Review — Eliminate orchestrator confirmation bias after fixes by restarting the full review with isolated sub-agents where applicable.

Why: The main agent knows what it (or $feature-implement) just fixed and rationalizes findings accordingly. A fresh sub-agent has ZERO memory, re-reads from scratch, and catches what the main agent dismissed. Sub-agent bias is mitigated by (1) fresh context, (2) verbatim protocol injection, (3) main agent not filtering the report.

When: ONLY after a validated-finding fix cycle. A review round that finds zero issues ENDS the loop — do NOT spawn a confirmation sub-agent. A review round that finds issues triggers: validate findings → fix → full review restart from the first phase.

How:

Start a NEW full review invocation/task breakdown; when that protocol calls for agents, spawn NEW spawn_agent tool calls — use code-reviewer agent_type for code reviews, general-purpose for plan/doc/artifact reviews

Inject ALL required review protocols VERBATIM into the prompt — see SYNC:review-protocol-injection for the full list and template. Never reference protocols by file path; AI compliance drops behind file-read indirection (see SYNC:shared-protocol-duplication-policy)

Sub-agent re-reads ALL target files from scratch via its own tool calls — never pass file contents inline in the prompt

Sub-agent writes structured report to plans/reports/{review-type}-round{N}-{date}.md

Main agent reads the report, integrates findings into its own report, DOES NOT override or filter

Rules:

SKIP fresh sub-agent when the prior full review found zero issues (no fixes = nothing new to verify)

NEVER skip the full review restart after a fix cycle — every fix invalidates the prior verdict

NEVER reuse a sub-agent across rounds — every fresh round spawns a NEW spawn_agent call

Continue until a complete full review pass has zero findings; if the same blocker repeats 3 times with no progress, escalate via a direct user question

Track iteration count and repeated blockers in conversation context (session-scoped, no persistent files)

Review Protocol Injection — Every fresh sub-agent review prompt MUST embed 11 protocol blocks VERBATIM. The template below has ALL 11 bodies already expanded inline. Copy the template wholesale into the Agent call's prompt field at runtime, replacing only the {placeholders} in Task / Round / Reference Docs / Target Files / Output sections with context-specific values. Do NOT touch the embedded protocol sections.

Why inline expansion: Placeholder markers would force file-read indirection at runtime. AI compliance drops significantly behind indirection (see SYNC:shared-protocol-duplication-policy). Therefore the template carries all 11 protocol bodies pre-embedded.

Subagent Type Selection

code-reviewer — for code reviews (reviewing source files, git diffs, implementation)
general-purpose — for plan / doc / artifact reviews (reviewing markdown plans, docs, specs)

Canonical Agent Call Template (Copy Verbatim)

spawn_agent({
  description: "Fresh Round {N} review",
  agent_type: "code-reviewer",
  prompt: `
## Task
{review-specific task — e.g., "Review all uncommitted changes for code quality" | "Review plan files under {plan-dir}" | "Review integration tests in {path}"}

## Round
Round {N}. You have ZERO memory of prior rounds. Re-read all target files from scratch via your own tool calls. Do NOT trust anything from the main agent beyond this prompt.

## Protocols (follow VERBATIM — these are non-negotiable)

### Spec ↔ Tests ↔ Code Triangulation
DO THIS FIRST — before any per-protocol check below. The review target is the WHOLE PACKAGE, not the diff alone: load the behavior's spec (§3 ACs / §4 BRs / §8 TCs), its tests, and the changed code TOGETHER, and reason about their mutual consistency BEFORE judging any one in isolation.
1. Locate all three faces: the Feature Spec section(s) governing the changed behavior, the tests that guard it, and the production code that implements it. A missing face is itself a finding (SPEC-GAP / TEST-GAP / DEAD-SPEC).
2. Triangulate pairwise — every disagreement is a finding; classify which face is wrong:
   - code vs spec: behavior the code does that no §3/§4/§8 rule describes → CODE-EXTRA or SPEC-STALE; a [HARD] §4 rule or §5 invariant with no enforcing code path → CODE-WRONG.
   - tests vs spec: a §8 TC with no test, or a test asserting behavior no TC/rule names → TEST-GAP or SPEC-SILENT.
   - tests vs code: a changed code path with no covering test → TEST-GAP; a test that still passes against a deliberately broken invariant → WEAK-TEST (apply the mutation thinking in Bug Detection).
3. Hidden-rule capture: any invariant the code enforces but the spec never states (SPEC-SILENT) MUST be surfaced as a finding to add into §3/§4/§8 AND guarded with a test — the enrichment loop, never a silent pass.
4. Only after the three faces agree — or every disagreement is logged as a finding — proceed to the per-protocol checks below; when enrichment adds spec/test content, re-review the package against the enriched spec.
NEVER mark review PASS while any spec/test/code face disagrees without a logged finding. The diff is the entry point; the package is the unit of judgment.

### Evidence-Based Reasoning
Speculation is FORBIDDEN. Every claim needs proof.
1. Cite file:line, grep results, or framework docs for EVERY claim
2. Declare confidence: >80% act freely, 60-80% verify first, <60% DO NOT recommend
3. Cross-service validation required for architectural changes
4. "I don't have enough evidence" is valid and expected output
BLOCKED until: Evidence file path (file:line) provided; Grep search performed; 3+ similar patterns found; Confidence level stated.
Forbidden without proof: "obviously", "I think", "should be", "probably", "this is because".
If incomplete → output: "Insufficient evidence. Verified: [...]. Not verified: [...]."

### Bug Detection
MUST check categories 1-4 for EVERY review. Never skip.
1. Null Safety: Can params/returns be null? Are they guarded? Optional chaining gaps? .find() returns checked?
2. Boundary Conditions: Off-by-one (< vs <=)? Empty collections handled? Zero/negative values? Max limits?
3. Error Handling: Try-catch scope correct? Silent swallowed exceptions? Error types specific? Cleanup in finally?
4. Resource Management: Connections/streams closed? Subscriptions unsubscribed on destroy? Timers cleared? Memory bounded?
5. Concurrency (if async): Missing await? Race conditions on shared state? Stale closures? Retry storms?
6. Stack-Specific: Check the configured language/runtime pitfalls and framework-specific failure modes discovered from local code.
Classify: CRITICAL (crash/corrupt) → FAIL | HIGH (incorrect behavior) → FAIL | MEDIUM (edge case) → WARN | LOW (defensive) → INFO.

### Design Patterns Quality
Priority checks for every code change:
1. DRY via OOP: Same-suffix classes (*Entity, *Dto, *Service) MUST share base class. 3+ similar patterns → extract to shared abstraction.
2. Right Responsibility: Logic in LOWEST layer (Entity > Domain Service > Application Service > Controller). Never business logic in controllers.
3. SOLID: Single responsibility (one reason to change). Open-closed (extend, don't modify). Liskov (subtypes substitutable). Interface segregation (small interfaces). Dependency inversion (depend on abstractions).
4. After extraction/move/rename: Grep ENTIRE scope for dangling references. Zero tolerance.
5. YAGNI gate: NEVER recommend patterns unless 3+ occurrences exist. Don't extract for hypothetical future use.
Anti-patterns to flag: God Object, Copy-Paste inheritance, Circular Dependency, Leaky Abstraction.

### Logic & Intention Review
Verify WHAT code does matches WHY it was changed.
1. Change Intention Check: Every changed file MUST serve the stated purpose. Flag unrelated changes as scope creep.
2. Happy Path Trace: Walk through one complete success scenario through changed code.
3. Error Path Trace: Walk through one failure/edge case scenario through changed code.
4. Acceptance Mapping: If plan context available, map every acceptance criterion to a code change.
5. Tests Verify Intent: For test/spec changes, verify tests name the protected business rule or invariant and would fail if that intent breaks.
6. Migration Test Exclusion: Do not write tests for migration code. Schema/data migrations are one-time execution paths, not core application logic.
NEVER mark review PASS without completing both traces (happy + error path).

### Test Spec Verification
Map changed code to test specifications.
1. Identify the project's test/spec format from existing docs, test-case files, BDD feature files, or spec folders.
2. Every changed code path MUST map to a corresponding test case/spec (or flag as "needs test case").
3. New functions/endpoints/handlers → flag for test spec creation.
4. Migration files are excluded from test/spec creation; schema/data migrations are one-time execution paths, not core application logic.
5. If spec evidence fields exist, verify they point to actual code (file:line, not stale references).
6. Verify each meaningful test case names the business intent/invariant; flag behavior-only cases that only mirror implementation details.
7. Auth/data changes → verify corresponding authorization and data-state test cases exist.
8. If no specs exist for a changed path → log the gap and recommend the project's test-spec workflow.
NEVER skip test mapping. Untested code paths are the #1 source of production bugs.

### Behavioral Delta Matrix
MANDATORY for any bugfix review. Produce input-state × pre-fix × post-fix × delta table BEFORE writing verdict.
- Minimum 3 rows; include at least one row OUTSIDE the original bug report.
- Any "REGRESSION" delta → review returns FAIL until a preservation test is added.
- Narrative descriptions do NOT substitute for the matrix.
Example rows (external-record sync fix):
| Input                 | Pre-fix | Post-fix                  | Delta      |
| --------------------- | ------- | ------------------------- | ---------- |
| Record exists (valid) | Reused  | Always recreated → orphan | REGRESSION |
| Record missing (404)  | Error   | Recreated                 | Fixed      |

### Fix-Layer Accountability
NEVER fix at the crash site. Trace the full flow, fix at the owning layer. The crash site is a SYMPTOM, not the cause.
MANDATORY before ANY fix:
1. Trace full data flow — Map the complete path from data origin to crash site across ALL layers (storage → backend → API → frontend → UI). Identify where bad state ENTERS, not where it CRASHES.
2. Identify the invariant owner — Which layer's contract guarantees this value is valid? Fix at the LOWEST layer that owns the invariant, not the highest layer that consumes it.
3. One fix, maximum protection — If fix requires touching 3+ files with defensive checks, you are at the wrong layer — go lower.
4. Verify no bypass paths — Confirm all data flows through the fix point. Check for direct construction skipping factories, clone/spread without re-validation, raw data not wrapped in domain models, mutations outside the model layer.
BLOCKED until: Full data flow traced (origin → crash); Invariant owner identified with file:line evidence; All access sites audited (grep count); Fix layer justified (lowest layer that protects most consumers).
Anti-patterns (REJECT): "Fix it where it crashes" (crash site ≠ cause site, trace upstream); "Add defensive checks at every consumer" (scattered defense = wrong layer); "Both fix is safer" (pick ONE authoritative layer).

### Rationalization Prevention
AI skips steps via these evasions. Recognize and reject:
- "Too simple for a plan" → Simple + wrong assumptions = wasted time. Plan anyway.
- "I'll test after" → RED before GREEN. Write/verify test first.
- "Already searched" → Show grep evidence with file:line. No proof = no search.
- "Just do it" → Still need task tracking. Skip depth, never skip tracking.
- "Just a small fix" → Small fix in wrong location cascades. Verify file:line first.
- "Code is self-explanatory" → Future readers need evidence trail. Document anyway.
- "Combine steps to save time" → Combined steps dilute focus. Each step has distinct purpose.

### Graph-Assisted Investigation
MANDATORY when .code-graph/graph.db exists.
HARD-GATE: MUST run at least ONE graph command on key files before concluding any investigation.
Pattern: Grep finds files → trace --direction both reveals full system flow → Grep verifies details.
- Investigation/Scout: trace --direction both on 2-3 entry files
- Fix/Debug: callers_of on buggy function + tests_for
- Feature/Enhancement: connections on files to be modified
- Code Review: tests_for on changed functions
- Blast Radius: trace --direction downstream
CLI: python .claude/scripts/code_graph {command} --json. Use --node-mode file first (10-30x less noise), then --node-mode function for detail.

### Understand Code First
HARD-GATE: Do NOT write, plan, or fix until you READ existing code.
1. Search 3+ similar patterns (grep/glob) — cite file:line evidence.
2. Read existing files in target area — understand structure, base classes, conventions.
3. Run python .claude/scripts/code_graph trace <file> --direction both --json when .code-graph/graph.db exists.
4. Map dependencies via connections or callers_of — know what depends on your target.
5. Write investigation to .ai/workspace/analysis/ for non-trivial tasks (3+ files).
6. Re-read analysis file before implementing — never work from memory alone.
7. NEVER invent new patterns when existing ones work — match exactly or document deviation.
BLOCKED until: Read target files; Grep 3+ patterns; Graph trace (if graph.db exists); Assumptions verified with evidence.

## Reference Docs (READ before reviewing)
- `.claude/docs/development-rules.md` — canonical development rules, code-quality guidelines, and pre-commit checklist
- docs/project-reference/code-review-rules.md
- {skill-specific reference docs — e.g., integration-test-reference.md for integration-test-review; backend-patterns-reference.md for backend reviews; frontend-patterns-reference.md for frontend reviews}

## Target Files
{explicit file list OR "run git diff to see uncommitted changes" OR "read all files under {plan-dir}"}

## Output
Write a structured report to plans/reports/{review-type}-round{N}-{date}.md with sections:
- Status: PASS | FAIL
- Issue Count: {number}
- Critical Issues (with file:line evidence)
- High Priority Issues (with file:line evidence)
- Medium / Low Issues
- Cross-cutting findings

Return the report path and status to the main agent.
Every finding MUST have file:line evidence. Speculation is forbidden.
`
})

Rules

DO copy the template wholesale — including all 11 embedded protocol sections
DO replace only the {placeholders} in Task / Round / Reference Docs / Target Files / Output sections with context-specific content
DO choose code-reviewer agent_type for code reviews and general-purpose for plan / doc / artifact reviews
DO NOT paraphrase, summarize, or skip any protocol section
DO NOT pass file contents inline — the sub-agent reads via its own tool calls so it has a fresh context
DO NOT reference protocols by file path or tag name — the bodies are already embedded above
DO NOT introduce placeholder markers for the protocols — they must stay literally expanded

Logic & Intention Review — Verify WHAT code does matches WHY it was changed.

Change Intention Check: Every changed file MUST ATTENTION serve the stated purpose. Flag unrelated changes as scope creep.

Happy Path Trace: Walk through one complete success scenario through changed code

Error Path Trace: Walk through one failure/edge case scenario through changed code

Acceptance Mapping: If plan context available, map every acceptance criterion to a code change

Tests Verify Intent: For test/spec changes, verify tests name the protected business rule or invariant and would fail if that intent breaks.

Migration Test Exclusion: Do not write tests for migration code. Schema/data migrations are one-time execution paths, not core application logic.

NEVER mark review PASS without completing both traces (happy + error path).

Bug Detection — MUST ATTENTION check categories 1-4 for EVERY review. Never skip.

Null Safety: Can params/returns be null? Are they guarded? Optional chaining gaps? .find() returns checked?

Boundary Conditions: Off-by-one (< vs <=)? Empty collections handled? Zero/negative values? Max limits?

Error Handling: Try-catch scope correct? Silent swallowed exceptions? Error types specific? Cleanup in finally?

Resource Management: Connections/streams closed? Subscriptions unsubscribed on destroy? Timers cleared? Memory bounded?

Concurrency (if async): Missing await? Race conditions on shared state? Stale closures? Retry storms?

Stack-Specific: Check the configured language/runtime pitfalls and framework-specific failure modes discovered from local code.

Classify: CRITICAL (crash/corrupt) → FAIL | HIGH (incorrect behavior) → FAIL | MEDIUM (edge case) → WARN | LOW (defensive) → INFO

Test Spec Verification — Map changed code to test specifications.

Identify the project's test/spec format from existing docs, test-case files, BDD feature files, or spec folders.

Every changed code path MUST ATTENTION map to a corresponding test case/spec (or flag as "needs test case")

New functions/endpoints/handlers → flag for test spec creation

Migration files are excluded from TC/test creation; schema/data migrations are one-time execution paths, not core application logic.

If spec evidence fields exist, verify they point to actual code (file:line, not stale references)

Verify each meaningful test case names the business intent/invariant; flag behavior-only cases that only mirror implementation details.

Auth/data changes → verify corresponding authorization and data-state test cases exist.

If no specs exist for a changed path → log the gap and recommend the project's test-spec workflow.

NEVER skip test mapping. Untested code paths are the #1 source of production bugs.

Integration Test Sync Check — Verify changed business logic files have corresponding tests.

From changed files → identify business logic files: handlers, commands, queries, services, controllers, resolvers, event processors. Naming varies by stack — infer from project conventions (e.g., *Service.*, *Handler.*, *Controller.*, *Command.*, *Query.*, *Resolver.*). Exclude migration files: schema/data migrations are one-time execution paths, not core application logic.

For each identified file → search for a corresponding test file. Infer test naming from existing tests in the project (e.g., *.test.ts, *Tests.java, *_test.py, *.spec.js, *Tests.cs). Check standard test directories (tests/, spec/, __tests__/, or adjacent test projects/packages).

If test EXISTS → check if test methods cover changed behavior (new methods/parameters/logic paths)

If test MISSING → MANDATORY: use a direct user question: "Business logic file {file} has no integration tests — run $integration-test before proceeding, or confirm tests already written?" Options: "Run $integration-test first" (Recommended) | "Tests already written/updated — proceed"

Severity: HIGH — missing tests for changed business logic MUST be surfaced to the user; do NOT silently flag and continue

Surface every business-logic change that lacks test coverage for an explicit a direct user question decision — never silently skip. — why: a silent skip ships untested business logic to production.

Translation Sync Check — Verify multilingual UI changes include translation updates.

Determine multilingual mode from project config: localization.enabled === true and supportedLocales.length > 1

Detect UI-facing file changes via extensions/path patterns (.ts, .tsx, .html, .css, .scss plus localization.uiPathPatterns when configured)

For multilingual UI changes, verify translation resource diffs exist (localization.translationFilePatterns when configured)

If translation updates are missing → MANDATORY: use a direct user question: "UI text changed in a multilingual project, but translation updates were not detected. Run translation sync now or proceed with explicit risk acceptance?" Options: "Run translation sync first" (Recommended) | "Proceed with explicit risk acceptance"

Severity: HIGH — no silent pass for multilingual UI text changes without explicit translation-sync decision

Do NOT silently skip. Multilingual UI text changes require explicit translation-sync confirmation.

Category Review Thinking — A thinking framework for reviewing any category of changed files. NOT a fixed checklist — derive concerns from domain knowledge; the examples are starting points only. Your knowledge of the category exceeds any list here — trust it.

Step 1 — Understand the category's role. What is this category responsible for in the overall system? What invariants must it uphold? What are its consumer contracts (who depends on it, what do they expect)?

Step 2 — Read project conventions for this category. Search for reference docs, style guides, ADRs, or READMEs specific to this area. Grep 3+ existing similar files — extract naming conventions, structural patterns, shared base classes. If no docs exist, derive conventions empirically from existing code.

Step 3 — Derive concerns from first principles. Apply all that are relevant; expand beyond this list based on the actual category:

Correctness: Does the logic match the intent? Trace happy path AND error path.

Boundary contracts: Are interfaces/APIs/events/protocols honored? No implicit coupling introduced?

Project conventions: Does new code follow the patterns found in Step 2? Evidence-confirmed, not assumed.

Security: Auth enforced at every entry point? Input validated at boundaries? No secrets in the diff?

Performance: Unbounded operations? N+1 patterns? Blocking calls in async context? Unindexed queries?

Maintainability: DRY? Single responsibility? Complexity within reason? Names reveal intent?

Test coverage: Are the changed paths covered by tests? Are existing tests still valid after the change?

Documentation: Do related docs, specs, or READMEs reflect the changes?

Step 4 — Create sub-tasks and execute. For each identified concern: create a task tracking sub-task, work through it with file:line evidence, mark done. No findings without proof.

Illustrative concern examples by category type (not exhaustive — trust your knowledge beyond this):

Server-side logic: handler/service structure conventions, validation layer placement, side-effect isolation, cross-service boundary enforcement, data-access layer separation, error propagation strategy

Client-side logic: component lifecycle management, resource cleanup (subscriptions, listeners, timers), state management patterns, API integration layer separation, reactive stream composition

Data/Schema: migration reversibility (rollback script), lock impact on table volume, backfill idempotency, index coverage for query patterns, deployment ordering

Configuration: present in ALL environments? No secrets in diff? App fails fast if config missing (not silently null)? Documented in setup guide?

Infrastructure: dev/prod parity? No hardcoded dev values (localhost, debug flags)? Pinned image/dependency versions? CI/CD secret requirements documented?

Styles/Assets: follows project naming conventions? Uses design variables/tokens (no hardcoded magic values)? Correct scope (no global side effects from component styles)?

Documentation: accurate? Links valid? Examples still match current code/behavior? Covers new scenarios?

Tests: assertions verify specific outcomes (not just "no exception")? Idempotent (repeatable N times)? Covers edge cases, not just happy path?

Security artifacts: all code paths reach the gate? Negative tests exist (unauthorized denied)? Both enforcement AND display control updated?

Build/Tooling: rule changes apply consistently? No exceptions that silently swallow violations? Impact on CI runtime documented?

Graph-Assisted Investigation — MANDATORY when .code-graph/graph.db exists.

HARD-GATE: MUST ATTENTION run at least ONE graph command on key files before concluding any investigation.

Pattern: Grep finds files → trace --direction both reveals full system flow → Grep verifies details

| Task | Minimum Graph Action | | ------------------- | -------------------------------------------- | | Investigation/Scout | trace --direction both on 2-3 entry files | | Fix/Debug | callers_of on buggy function + tests_for | | Feature/Enhancement | connections on files to be modified | | Code Review | tests_for on changed functions | | Blast Radius | trace --direction downstream |

CLI: python .claude/scripts/code_graph {command} --json. Use --node-mode file first (10-30x less noise), then --node-mode function for detail.

Nested Task Expansion Contract — For workflow-step invocation, the [Workflow] ... row is only a parent container; the child skill still creates visible phase tasks.

Call the current task list first. If a matching active parent workflow row exists, set nested=true and record parentTaskId; otherwise run standalone.

Create one task per declared phase before phase work. When nested, prefix subjects [N.M] $skill-name — phase.

When nested, link the parent with TaskUpdate(parentTaskId, addBlockedBy: [childIds]).

Orchestrators must pre-expand a child skill's phase list and link the workflow row before invoking that child skill or sub-agent.

Mark exactly one child in_progress before work and completed immediately after evidence is written.

Complete the parent only after all child tasks are completed or explicitly cancelled with reason.

Blocked until: the current task list done, child phases created, parent linked when nested, first child marked in_progress.

Project Reference Docs Gate — Run after task-tracking bootstrap and before target/source file reads, grep, edits, or analysis. Project docs override generic framework assumptions.

Identify scope: file types, domain area, and operation.

Required docs by trigger: always docs/project-reference/lessons.md; doc lookup docs-index-reference.md; review code-review-rules.md; backend/CQRS/API backend-patterns-reference.md; domain/entity domain-entities-reference.md; frontend/UI frontend-patterns-reference.md; styles/design scss-styling-guide.md + design-system/design-system-canonical.md; integration tests integration-test-reference.md; E2E e2e-test-reference.md; feature docs/specs feature-spec-reference.md + spec-system-reference.md + spec-principles.md; behavior/public-contract/spec-test-code sync workflow-spec-test-code-cycle-reference.md; derived spec index/ERD/reimplementation guides spec-system-reference.md + source Feature Specs under docs/specs/; architecture/new area project-structure-reference.md.

Read every required doc. If docs/project-config.json, the docs index, lessons.md, CLAUDE.md, AGENTS.md, or any task-required reference doc is missing or stale, auto-run $project-init or the narrow lower-level route ($project-config, $docs-init, $scan-all, $scan --target=<key>, $claude-md-init) before ordinary project-specific work. If Codex mirrors or AGENTS.md are missing/stale, ask the user to run $sync-codex; do not auto-run it.

Before target work, state: Reference docs read: ... | Not applicable: ....

Ready when: scope evaluated, required docs checked/read or setup route completed, lessons.md confirmed, citation emitted.

Task Tracking & External Report Persistence — Bootstrap this before execution; then run project-reference doc prefetch before target/source work.

Create a small task breakdown before target file reads, grep, edits, or analysis. On context loss, inspect the current task list first.

Mark one task in_progress before work and completed immediately after evidence; never batch transitions.

For plan/review work, create plans/reports/{skill}-{YYMMDD}-{HHmm}-{slug}.md before first finding.

Append findings after each file/section/decision and synthesize from the report file at the end.

Final output cites Full report: plans/reports/{filename}.

Blocked until: task breakdown exists, report path declared for plan/review work, first finding persisted before the next finding.

Source/test drift check. For coding, fix, debug, investigation, test, or review work: when source behavior changes, inspect affected unit/integration/E2E tests and decide from evidence whether tests should change to match intended behavior or the source change is an unintended bug to fix. Do not write tests for migration code; schema/data migrations are one-time execution paths, not core application logic.

Spec drift adjudication (code-wrong vs spec-stale). Whenever changed behavior diverges from a canonical Feature Spec (business rule, acceptance criterion, flow, state transition, or §8 TC under docs/specs/), you MUST NOT silently pick a side. Adjudicate per shared/sdd-artifact-contract.md → Drift Gates:

Detect — compare the change against the spec's documented intent. No divergence → record Spec in sync and move on.

Classify the divergence:

CODE-WRONG — the spec correctly states intended behavior and the change violates it → BLOCKING finding; fix the code/test against intended behavior (write/adjust a regression TC first).

SPEC-STALE — the change is the new intended behavior and the spec now documents the old/wrong behavior → update the spec FIRST via $spec [mode=update], then sync $spec [mode=tests] + $spec [mode=sync].

AMBIGUOUS — intended behavior is unclear → a direct user question (or the canonical spec owner) before editing either side.

SPEC-SILENT — the code correctly enforces an invariant/behavior that NO canonical spec artifact (§3 AC, §4 BR, §5 invariant, §8 TC) states → not drift but an UNWRITTEN rule discovered by review. ENRICH the spec via the Invariant Harvest pass ($spec [mode=sync] direction=harvest → spec/references/sync.md): prove it is always-true (≥2 enforcement points or a rejecting guard), express it as a universally-quantified property, then add the rule to §4 (or §3/§5) AND a §8 TC via $spec [update] + $spec [mode=tests] and add the guarding test. A discovered invariant left only in code (or only in tests) is INCOMPLETE — this is the highest-value capture (the rule nobody wrote down).

Never normalize drift just because code/tests are green — green can encode the drift itself. Reconcile to canonical intent, never to whichever side currently passes.

A behavior-changing review/implementation that leaves a spec divergence unadjudicated is INCOMPLETE; an unwritten-but-enforced invariant left uncaptured (no §4/§8 entry) is equally INCOMPLETE.

AI Mistake Prevention — Failure modes to avoid on every task:

Re-read files after context changes. Context compaction, resume, or long-running work can make memory stale; verify current files before acting. Verify generated content against source evidence. AI hallucinates APIs, names, claims, and document facts. Check the relevant source before documenting or referencing. Check downstream references before deleting or renaming. Removing an artifact can stale docs, generated mirrors, configs, and callers; map references first. Trace the full impact chain after edits. Changing a definition can miss derived outputs and consumers. Follow the affected chain before declaring done. Verify ALL affected outputs, not just the first. One green check is not all green checks; validate every output surface the change can affect. Assume existing values are intentional — ask WHY before changing. Before changing a constant, limit, flag, wording, or pattern, read nearby context and history. Surface ambiguity before acting — don't pick silently. Multiple valid interpretations require an explicit question or stated assumption with risk. Keep shared guidance role-relevant. Universal guidance must help every receiving skill or agent; code-specific obligations belong only in code-specific protocols.

Severity Rubric — Classify every finding by consequence, not by how easy it is to fix. One scale across all reviews so a "High" means the same thing everywhere.

| Severity | Action | Definition | | -------- | ----------- | ------------------------------------------------------------------------- | | CRITICAL | Block merge | Silent runtime failure, data corruption, validation bypass, security hole | | HIGH | Must fix | Incorrect behavior, invariant gap, architectural violation | | MEDIUM | Should fix | Design debt, maintainability, likely future bug | | LOW | Nice to fix | Convention, documentation, minor clarity |

Score-based skills map their numeric scale onto these tiers — do not invent a parallel vocabulary:

0-2 criterion scoring (e.g. production-readiness-review): 0 = CRITICAL/HIGH (criterion unmet, blocks production readiness), 1 = MEDIUM (partial, should fix), 2 = pass (no finding).

Two-axis scoring (e.g. performance-review, impact × likelihood): map the resulting cell to the nearest tier — high-impact + high-likelihood → CRITICAL/HIGH; low-impact OR low-likelihood → MEDIUM/LOW.

A finding's tier drives the gate: CRITICAL/HIGH must be resolved or explicitly accepted by the owner before PASS; MEDIUM/LOW may ship with a tracked follow-up.

IMPORTANT MUST ATTENTION search 3+ existing patterns and read code BEFORE any modification. Run graph trace when graph.db exists.

MANDATORY IMPORTANT MUST ATTENTION cite file:line evidence for every claim. Confidence >80% to act, <60% = do NOT recommend.

IMPORTANT MUST ATTENTION check DRY via OOP, right responsibility layer, SOLID. Grep for dangling refs after moves.

IMPORTANT MUST ATTENTION apply complexity prevention — one business change = one code change. Flag change amplification (>3 edit sites for future change), scattered type-switches, anemic models, primitive obsession, leaked technology through abstractions, shallow modules, un-extracted utility logic (paging/datetime/string/retry → helpers), and logic in the wrong higher layer (downshift to callee/entity/VM). Don't rationalize silent duplication with pure YAGNI.

IMPORTANT MUST ATTENTION run at least ONE graph command on key files when graph.db exists. Pattern: grep → trace → verify.

IMPORTANT MUST ATTENTION verify WHAT code does matches WHY it changed. Trace happy + error paths.

IMPORTANT MUST ATTENTION check null safety, boundaries, error handling, resource management for every review.

IMPORTANT MUST ATTENTION map changed code paths to test cases. Flag untested paths.

IMPORTANT MUST ATTENTION check changed logic files for matching tests. Surface missing tests via a direct user question — mandatory, not advisory.

IMPORTANT MUST ATTENTION for multilingual UI text changes, verify translation updates. If missing, require explicit user decision via a direct user question.

MUST ATTENTION apply critical + sequential thinking — every claim needs appropriate traced evidence (file:line for repo/code claims; source URL or artifact section for research, product, content, and docs claims); confidence >80% to act, <60% DO NOT recommend. Anti-hallucination: never present guess as fact, admit uncertainty freely, cross-reference independently, stay skeptical of own confidence.

MUST ATTENTION apply sequential-thinking — multi-step Thought N/M, REVISION/BRANCH/HYPOTHESIS markers, confidence % closer; see $sequential-thinking skill.

MUST ATTENTION apply AI mistake prevention — verify generated content against evidence, trace downstream references before deleting or renaming, verify all affected outputs, re-read files after context loss, and surface ambiguity before acting.

MANDATORY Bootstrap task tracking before target work; transition one task at a time.
MANDATORY Persist plan/review findings to plans/reports/ incrementally and synthesize from disk.

MANDATORY After task-tracking bootstrap and before target/source work, read required project-reference docs and cite Reference docs read: ....
MANDATORY Always include lessons.md; project conventions override generic defaults.
MANDATORY If project config, root instruction files, or any required reference doc is missing or stale, auto-run $project-init or the narrow lower-level route before ordinary project-specific work.

IMPORTANT MUST ATTENTION debugger trace gate: for non-trivial bug/fix/investigation/review work, start at the observed final output and trace backward through reader -> storage/projection -> writer -> consumer/job -> producer/trigger. Enumerate all feeder paths and hypotheses before fixing. BLOCKED until trace, hypothesis matrix, owning fix layer, and forward convergence proof exist.

MANDATORY Parent workflow rows do not replace child phase tracking; expand phases and link the parent when nested.
MANDATORY Orchestrators pre-expand child skill phases before invocation; use [N.M] $skill-name — phase prefixes and one-in_progress discipline.

MANDATORY Resolve the active Goal Contract BEFORE work (active plan goal.md → plans/goals/{YYMMDD-HHmm}-{slug}/goal.md → create from current request) and read saved success criteria before editing.
MANDATORY Append iteration evidence after execution; emit a Goal Satisfaction matrix (PASS/FAIL/BLOCKED) before reporting PASS; loop on validated FAIL; escalate repeated no-progress or blockers. NEVER store secrets in goal files.

MANDATORY Large changeset → batch by size cap (≤8 files OR ≤2000 diff-lines), one parallel sub-agent per batch; never review many files one-by-one.
MANDATORY > 6 categories OR > 40 files → add the hierarchical synthesis tier; each concern-synthesizer emits cross-concern interaction candidates and the orchestrator runs the cross-concern pass before concluding.

MANDATORY Classify findings Critical/High/Medium/Low by consequence; Critical/High block PASS until fixed or owner-accepted.
MANDATORY Score-based skills (sre 0-2, perf two-axis) map onto the same four tiers — no parallel severity vocabulary.

MANDATORY Derive review categories from file language + directory semantics + change nature; create a sub-task per category.
MANDATORY Derive each category's concerns from first principles with file:line evidence — never a fixed checklist.

Closing Reminders

IMPORTANT MUST ATTENTION Goal: Ensure every reviewed change is defect-free, evidence-backed, convention-aligned, and synchronized with required tests/docs before handoff; when code files changed, also prove the code stays easy to change.

IMPORTANT MUST ATTENTION — Protocols in force (concise digest of the SYNC/shared blocks this skill carries; each line is a signpost to its canonical body above — MUST ATTENTION honor the full block, NEVER act on the digest alone):

Systematic Batching: ≥10 files → size-capped parallel batches.
Debugger Trace: bug/fix → End→Start backward trace gate.
Critical Thinking: every claim traced; confidence >80% to act.
Sequential Thinking: multi-step Thought N/M with markers.
Understand Code First: grep 3+, read code before changes.
Design Patterns Quality: DRY/SOLID, right responsibility layer.
Complexity Prevention: one business change = one code change.
Double Round-Trip Review: validate → fix → full re-review until clean.
Fresh Context Review: fresh zero-memory sub-agent after fixes.
Review Protocol Injection: embed 11 protocols verbatim into sub-agent prompts.
Logic And Intention Review: WHAT code does matches WHY changed.
Bug Detection: null, boundaries, error handling, resource cleanup.
Test Spec Verification: map changed paths to test cases.
Integration Test Sync: missing tests surface via a direct user question.
Translation Sync: multilingual UI text changes need translation updates.
Category Review Thinking: derive categories from file + change nature.
Graph-Assisted Investigation: run graph trace on key files.
Nested Task Creation: expand child phases, link parent.
Project Reference Docs: read required docs including lessons.md.
Task Tracking + Report: bootstrap tasks, persist findings incrementally.
Source/Test Drift: behavior change → reconcile affected tests.
Spec Drift: classify CODE-WRONG/SPEC-STALE/AMBIGUOUS/SPEC-SILENT, route fix.
AI Mistake Prevention: verify generated content against evidence, trace downstream references, verify all affected outputs, re-read after context loss, surface ambiguity.
Severity Rubric: classify Critical/High/Medium/Low by consequence.

[CRITICAL — TOP 3 RULES REPEATED]

MUST ATTENTION Phase 0 graph blast-radius FIRST — NEVER skip; informs entire review priority order

MUST ATTENTION findings follow the active ownership boundary. Standalone mode runs Phase 6 validate → Phase 7 fix → full $review-changes restart; inside $workflow-review-changes, stop after the report and hand findings to parent step 2, then parent steps 10-15 own plan/feature-implement/restart.

MUST ATTENTION task tracking ALL phases before starting; missing tests MUST surface via a direct user question

MANDATORY Nested Task Expansion Contract — when invoked inside a workflow, STILL expand internal phases via task tracking with [N.M] $review-changes — phase prefix and TaskUpdate(parentTaskId, addBlockedBy: [childIds]) linkage. Workflow row is container, not substitute.
MANDATORY break work into small todo tasks using task tracking BEFORE starting
MANDATORY validate decisions with user via a direct user question — NEVER auto-decide
MANDATORY add final review todo task to verify work quality
MANDATORY discover and READ project-specific reference docs before starting
MANDATORY Phase 0 graph blast-radius is FIRST step — NEVER skip it
MANDATORY any finding must be validated before fix; standalone mode invokes $why-review --validate-findings, while $workflow-review-changes parent mode stops after the report and delegates validation to parent step 2
MANDATORY after fixing validated findings in standalone mode, recursively invoke $review-changes again from Phase 0 with a brand-new task breakdown and review the full current diff, not only the fixed files; in parent mode, parent steps 10-15 own the fix plan, feature-implement, and full restart
MANDATORY continue validate → fix → full restart until one complete review invocation has zero findings; standalone mode executes that loop locally, parent mode reports findings to $workflow-review-changes for the loop
MANDATORY documentation staleness check is REQUIRED in every review — flag stale docs even if not auto-fixing
MANDATORY run the Phase 8 final $docs-update gate once the review/fix loop converges to zero findings — ALWAYS, unconditional, never skipped on a clean verdict (deferred only inside $workflow-review-changes to the parent's docs-update step) — why: a passing code review still leaves docs stale until docs-update reconciles every impacted doc against the actual changes; a clean review with skipped docs-update is INCOMPLETE
MANDATORY run the Phase 3.5 Code-Simplifier Optimization gate whenever the diff includes code files — invoke $code-simplifier (report mode) over the changed code files, record its clarity/consistency/maintainability findings in the report, and route them through Phase 6 validation → Phase 7 fix; skip ONLY for docs-only diffs (record the skip reason) — why: correctness review proves it works, the simplifier gate proves it stays cheap to change, and the step is silently dropped without an anchored reminder
MANDATORY run the Phase 3.7 Integration-Test-Review Coverage Gate whenever the diff includes behavior-bearing code — invoke $integration-test-review over the FULL change set (production code AND tests); its Gate 7 must map every behavior change to a covering test (integration-first; unit fallback justified) and a spec TC, and every GAP/SPEC-GAP becomes a finding for Phase 6 validation → Phase 7 fix (GAP fix = write the missing test); skip ONLY for docs-only diffs with recorded reason, and defer to the parent's dedicated $integration-test-review step inside $workflow-review-changes — why: a correct-looking change with no covering test or stale spec ships unprotected behavior; the pairing check alone proves file names, not coverage
MANDATORY missing tests for changed business logic MUST surface to user via a direct user question — NOT silently logged
MANDATORY run the Phase 6 Why-Review Findings Validation Gate whenever ANY finding exists in standalone mode — invoke the $why-review skill through the skill invocation with --validate-findings (terminal mode, same session; an ACTUAL skill call — re-reading the cited lines yourself or any inline/manual self-validation does NOT satisfy this gate). The moment the first finding is recorded, register [Review Phase 6] Why-review findings validation gate — invoke $why-review via task tracking so it is never forgotten. Verify every finding is correct, proof-backed, reasonable, and best-practice; RE-DO it ONLY if it surfaces finding issues or enhancement opportunities (max 2 re-dos, then escalate via a direct user question); then Phase 7 fixes validated findings and restarts this skill. Inside $workflow-review-changes, do not run Phase 6/7 locally; parent step 2 and steps 10-15 own those gates.
MANDATORY follow declared step order; NEVER skip, reorder, or merge steps without explicit user approval
MANDATORY every skipped step includes explicit reason; every completed step includes concise evidence
MANDATORY Report-driven — write every finding to plans/reports/code-review-{date}-{slug}.md incrementally as each file/dimension is reviewed, NOT in one final batch — why: the report is external memory AND the deliverable; findings held only in context are lost on compaction
MANDATORY Evidence Gate — back every claim, finding, and recommendation with file:line proof or a traced call chain plus a confidence read (>80% act, 60-80% verify first, <60% DO NOT recommend); speculation is FORBIDDEN output and "insufficient evidence" is a valid verdict — why: an unproven finding wastes the author's time and erodes trust in the whole report
MANDATORY Convention-before-flag — grep 3+ existing examples before flagging ANY pattern, naming, or style violation; codebase convention wins over textbook rules — why: flagging against a rule the project never adopted manufactures false positives
MANDATORY Enforce YAGNI / KISS / DRY / Clean Code on changed code — flag speculative generality, needless complexity, and duplicated knowledge; 3+ similar patterns = MANDATORY extraction, 2+ same-kind violations = a structural finding (not individual nits) — why: these are the day-to-day forms of the Easy-to-Change success metric
MANDATORY Spec Drift Adjudication runs for EVERY behavior-changing review (not only post-bugfix) — classify each divergence CODE-WRONG / SPEC-STALE / AMBIGUOUS / SPEC-SILENT and route the owed fix (CODE-WRONG → fix code + regression TC; SPEC-STALE/SPEC-SILENT → $spec [update] + $spec [mode=tests]; AMBIGUOUS → a direct user question); record Spec in sync when none — why: passing code and tests can still silently normalize a spec violation
MANDATORY Dual-Feedback Ledger — every behavior-changing finding must feed BOTH the spec AND the tests; a blank or bare-"N/A" cell on either axis = FAIL (each N/A carries its reason inline) — why: fixing only the code leaves the spec or its TC stale and the gap reopens next change
MANDATORY Bugfix Debugger Trace Gate (§6.5) — for bugfix / regression / failed-verification / stale-output / behavior-changing fixes, FAIL the review unless an End→Start debugger trace, enumerated feeder paths, hypothesis matrix, lowest-owning-fix-layer justification, and forward-convergence + regression-test proof are all present — why: a fix without a trace patches the symptom site and the bug recurs
MANDATORY Fresh-Context Gate (Phase 3) is review-ONLY — it may ADD findings but NEVER fixes or validates them, NEVER re-reviews known findings, and does NOT run merely because Phase 7 restarted; any non-zero finding set flows straight to Phase 6 — why: re-reviewing known findings burns context without adding signal
MANDATORY Integrate sub-agent and synthesis findings RAW into the main report — read them, NEVER filter, soften, or override them — why: the main agent rationalizes away its own mistakes; the zero-memory sub-agent exists to catch exactly those
MANDATORY When the diff includes frontend/UI files, $review-changes OWNS the UI dimension — run $review-ui (prefer a ui-ux-designer sub-agent in the same parallel batch): long-content overflow, responsive flex sizing, z-index discipline, SCSS/BEM quality; skip only when no frontend files changed — why: $review-ui is never a separate workflow step, so a skipped UI pass ships UI defects unreviewed
MANDATORY M1-M6 Code-to-Spec Drift Gate (§8, BLOCKING) for any spec/feature-doc/PBI/story/test-spec the change touches or should sync — FAIL on introduced tech leakage (M1) / source-code-as-prose (M2) / broken logical-ID mapping (M3) / AC ambiguity (M4) / spec no longer rebuildable from artifact alone (M5); a FAIL must name the violated mandate ID + changed file:line; carriers ([Source:], **Evidence**, frontmatter, mermaid) are EXEMPT — why: passing an introduced M1-M5 violation makes the review itself defective
MANDATORY Architecture Boundary Check — for each changed file, read architectureRules.layerBoundaries from docs/project-config.json, match the file's layer, grep its imports, and BLOCK as Critical on any import from a cannotImportFrom layer; skip silently when architectureRules is absent — why: a layer-boundary breach compiles and ships silently while rotting the architecture
MANDATORY AI Agent Integrity Gate before reporting ANY work done — grep every removed name (0 dangling refs across ALL file types), ask WHY before changing an existing value, verify ALL affected outputs (one green ≠ all green), evaluate pattern fit before copying nearby code (same scope/lifetime/base class/constraints), and prove every new artifact is wired (registered, imported, reachable) — why: completion ≠ correctness, and an unverified "done" is an untested hypothesis
MANDATORY Large changeset (≥10 changed files) → Systematic Review Batching — categorize, size-cap batches (≤8 files OR ≤2000 diff-lines), fire one parallel sub-agent per batch, then reduce; >6 categories OR >40 files adds the hierarchical synthesis tier with the cross-concern interaction pass; ANNOUNCE any dropped/sampled scope — never review many files one-by-one and never let bounded coverage read as complete — why: a single agent's context silently truncates on big diffs, leaving files unreviewed

[BOTTOM REMINDER — WHY-REVIEW FINDINGS-VALIDATION GATE IS NON-NEGOTIABLE]

Same rule as the Top Reminder, repeated so it survives long contexts: ANY finding in standalone mode → you MUST invoke the $why-review skill (skill invocation, --validate-findings <report-path>) before any fix, docs-update, commit, or handoff. No inline/manual self-validation substitutes for the actual skill call. task tracking the [Review Phase 6] gate the moment the first finding lands so it can never be forgotten. Inside $workflow-review-changes, defer to parent step 2.

Anti-Rationalization:

| Evasion | Rebuttal | ----------------------------------------- | "Too simple for graph blast-radius" | "No findings, skip docs/tests" | "Finding is obvious, fix now" | "I already re-checked the lines myself" | "Only re-check fixed files" | "Sub-agent already reviewed" | "Already searched project conventions" | "DRY/SOLID requires this abstraction" | "Review found no bugs, skip simplifier" | "No test files changed, skip test | "Test file with matching name exists" | "Clean review, docs surely fine" | "Behavior changed but spec + tests pass" | "Bug fixed, ship it" | "Confident enough, no need to cite" | "It reads fine, looks correct" | "Frontend tweak, skip UI review" | "Nearby code does it this way, copy it" | "Spec prose names the class, harmless" | "Import compiles, layer is fine" | "Many files, I'll just read them all" | | ---------------------------------------------------------------------------------------------------------------------------------------------------- | | Phase 0 graph check sets risk order; run it or record graph unavailable. | | Clean verdict still needs proof that docs/tests were checked or explicitly not relevant. | | Invoke the $why-review skill (--validate-findings) first — an actual skill call, not inline self-validation; unvalidated findings are not fixes. | | Inline re-reading does NOT pass Phase 6. The gate requires a real $why-review skill invocation that returns a verdict. | | Fixes can interact with earlier changes; restart $review-changes from Phase 0 on the full diff. | | Main report must integrate raw findings and not override or filter them. | | Show 3+ file:line examples. No evidence means no search. | | Prove it lowers future change cost; otherwise it is ceremony, not quality. | | Bug-finding ≠ simplification. Run Phase 3.5 $code-simplifier on changed code files anyway. | review" | The review target is the CHANGE, not the test files. Phase 3.7 Gate 7 maps every behavior change to a covering test + spec TC. | | Name pairing ≠ coverage. Phase 3.7 $integration-test-review proves assertion-level coverage of the changed behavior. | | Clean code ≠ current docs. Run the Phase 8 $docs-update sweep before handoff — it is mandatory, not conditional. | | Passing ≠ in-sync. Run Spec Drift Adjudication AND the Dual-Feedback Ledger; classify the divergence and route the owed spec/test fix. | | A bugfix review FAILS without the §6.5 End→Start trace, feeder enumeration, hypothesis matrix, and regression proof. | | <80% confidence = verify first; no file:line = speculation, which is forbidden output. State what you traced and how. | | "Looks fine" is not proof. Trace one happy path + one error path and cite the lines before calling it correct. | | $review-changes owns the UI dimension — any frontend file in the diff triggers $review-ui; it is never a separate step to defer. | | Closest example ≠ matching preconditions. Prove the new context shares the same scope, lifetime, base class, and constraints before copying. | | Introduced tech/source-as-prose in spec narrative is an M1/M2 FAIL (§8). Name the mandate ID + changed file:line; carriers are the only exemption. | | Compiling ≠ allowed. Check architectureRules.layerBoundaries; a cannotImportFrom breach is a BLOCKING Critical, not a style nit. | | One context truncates silently on big diffs. ≥10 files → size-capped parallel batches + reduce; announce any dropped/sampled scope. |

[TASK-PLANNING] Break scope into small todo tasks before acting; maintain one in_progress; add final review todo.

Closing reminder — Easy to Change is the success metric. Every finding, test, refactor, and abstraction must answer one question: does this make the next change cheaper or more expensive? If it doesn't reduce future change cost, reject it. Coupling, hidden state, duplicated knowledge, and unclear intent are the real enemies — call them out by name.

IMPORTANT MUST ATTENTION graph blast-radius runs first when .code-graph/graph.db exists. IMPORTANT MUST ATTENTION every claim needs file:line proof; every stale docs/tests decision needs evidence. IMPORTANT MUST ATTENTION Goal: Ensure every reviewed change is defect-free, evidence-backed, convention-aligned, and synchronized with required tests/docs before handoff; when code files changed, also prove the code stays easy to change.

Hookless Prompt Protocol Mirror (Auto-Synced)

Source: .claude/.ck.json + .claude/skills/shared/sync-inline-versions.md (:full blocks) + .claude/scripts/lib/hookless-prompt-protocol.cjs

[WORKFLOW-EXECUTION-PROTOCOL] [BLOCKING] Workflow Execution Protocol — MANDATORY IMPORTANT MUST CRITICAL. Do not skip for any reason.

Generic portability boundary: Reusable skills and protocol text stay project-neutral; project-specific conventions are discovered from docs/project-config.json and docs/project-reference/. Apply shared AI-SDD from shared/sdd-artifact-contract.md. Read docs/project-config.json and docs/project-reference/docs-index-reference.md, then open the project reference docs named there. For spec, test-case, behavior-change, public-contract, or docs/specs/ work, route through the local spec docs named by the docs index: feature-spec-reference.md, spec-system-reference.md, spec-principles.md, and workflow-spec-test-code-cycle-reference.md when specs/tests/code must stay synchronized. If either file or a required reference doc is missing or stale, auto-run $project-init (or the narrow lower-level route such as $project-config, $docs-init, $scan-all, or $scan --target=<key>) before ordinary project-specific work. Any supported AI tool may execute when this shared context and local docs are available.

DETECT: If the prompt starts with an explicit slash skill/workflow command, execute it directly. Otherwise match the prompt against the workflow catalog and skill list.
ANALYZE: Choose the best option: execute directly, invoke a skill, activate a standard workflow, or compose a custom step combination.
AUTO-SELECT: Pick the best option yourself. Do not ask the user to choose between direct execution, skill, standard workflow, or custom workflow.
ACTIVATE: For a selected workflow, call $start-workflow <workflowId>; for a selected skill, invoke that skill; for a custom workflow, sequence custom steps directly; for direct execution, proceed with the task.
CREATE TASKS: task tracking for ALL workflow/skill/custom steps before execution when the selected path has multiple steps.
EXECUTE: Advance per the Workflow Step Advancement & Parallel Phases rule in your context instructions — model-driven; a sub-agent completion advances a step identically to an inline call; a parallel-phase group is an all-return barrier (advance only after ALL members return, never serialize it)

Shared AI-SDD Protocol Markers

Source: .claude/skills/shared/sync-inline-versions.md

SYNC:ai-sdd-artifact-contract

AI-SDD Artifact Contract — Shared spec-driven development rules stay portable and source-owned.

Keep reusable AI-SDD principles in .claude; put repository-specific paths, commands, owners, products, and formats in project config/reference docs.

Preserve cycle: spec -> plan -> tasks -> implement -> verify -> update spec/docs.

Trace every requirement or invariant through decision, task, TC/test, source evidence, and docs/spec update.

Treat code-to-spec extraction as reference-only until accepted by the canonical spec owner.

Any supported AI tool may plan, implement, review, or verify with synced context; using multiple tools is optional.

Update .claude source first, then sync generated mirrors; do not manually edit .agents, .codex, or AGENTS.md. — why: mirrors are generated artifacts; hand-edits are overwritten on the next sync

If docs/project-config.json, root instruction files, or a required project-reference doc is missing or stale, auto-run $project-init or the narrow lower-level route before ordinary project-specific work.

Active reference: shared/sdd-artifact-contract.md in the active skills root.

SYNC:ai-sdd-artifact-contract:reminder

MANDATORY Apply shared/sdd-artifact-contract.md; keep reusable AI-SDD in .claude and local rules in project docs.
MANDATORY Code-to-spec extraction is reference-only until canonical acceptance; any supported AI tool may execute with synced context.
MANDATORY Update .claude source before syncing generated mirrors; do not manually edit .agents, .codex, or AGENTS.md.
MANDATORY Missing or stale project config, root instruction files, or required reference docs route project-specific work through $project-init or the narrow setup route automatically. [TASK-PLANNING] [MANDATORY] BEFORE executing any workflow or skill step, create/update task tracking for all planned steps, then keep it synchronized as each step starts/completes.

[LESSON-LEARNED-REMINDER] [BLOCKING] Task Planning & Continuous Improvement — MANDATORY. Do not skip.

Break work into small tasks (task tracking) before starting. Add final task: "Analyze AI mistakes & lessons learned".

Extract lessons — ROOT CAUSE ONLY, not symptom fixes:

Name the FAILURE MODE (reasoning/assumption failure), not symptom — "assumed API existed without reading source" not "used wrong enum value".
Generality test: does this failure mode apply to ≥3 contexts/codebases? If not, abstract one level up.
Write as a universal rule — strip project-specific names/paths/classes. Useful on any codebase.
Consolidate: multiple mistakes sharing one failure mode → ONE lesson.
Recurrence gate: "Would this recur in future session WITHOUT this reminder?" — No → skip $learn.
Auto-fix gate: "Could $code-review/$code-simplifier/$security-review/$lint catch this?" — Yes → improve review skill instead.
BOTH gates pass → ask user to run $learn. [CRITICAL-THINKING-MINDSET] Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence >80% to act. Anti-hallucination principle: Never present guess as fact — cite sources for every claim, admit uncertainty freely, self-check output for errors, cross-reference independently, stay skeptical of own confidence — certainty without evidence root of all hallucination. AI Attention principle (Primacy-Recency): Put the 3 most critical rules at both top and bottom of long prompts/protocols so instruction adherence survives long context windows. Goal-driven execution: Define success criteria first, loop until verified, and stop only when observable checks pass. Tests verify intent: Tests must protect business rules/invariants and fail when the protected intent breaks, not only mirror current behavior.

Common AI Mistake Prevention (System Lessons)

Re-read files after context compaction. Edit requires prior Read in same context; compaction wipes read state. Re-read before editing.
Grep for old terms after bulk replacements. AI over-trusts find/replace completeness. Grep full repo after bulk edits for missed refs in docs/configs/catalogs.
Check downstream references before deleting. Deletions cascade doc/code staleness. Map referencing files before removal.
After memory loss, check existing state before creating new. Compaction wipes prior-work memory. Query current state to resume — never blindly duplicate.
Verify AI-generated content against actual code. AI hallucinates APIs, class names, method signatures. Grep to confirm existence before documenting/referencing.
Trace full dependency chain after edits. Changing a definition misses downstream consumers. Trace the full chain.
When renaming, grep ALL consumer file types. Some file types silently ignore missing refs (no compile error). Search code, templates, configs, generated files.
Trace ALL code paths when verifying correctness. Code existing ≠ code executing. Trace early exits, error branches, conditional skips — not just happy path.
Update docs that embed canonical data when source changes. Docs inlining derived data (workflows, schemas, configs) go stale silently. Update all embedding docs alongside source.
Verify sub-agent results after context recovery. Background agents may finish while parent compacted — grep-verify output, don't trust assumed completion.
Cross-check full target list against sub-agent assignments. Parallel sub-agents by category miss boundary items. Reconcile union of assignments against target list before proceeding.
Sub-agents inherit knowledge only from their agent .md definition — use custom agent types, not built-in Explore. Tool adoption = permission + knowledge + enforcement (numbered workflow step).
Persist sub-agent findings incrementally, not as a final batch. Long sub-agents hit cutoffs before final write — findings lost. Instruct append-per-section to report file.
When debugging, ask "whose responsibility?" before fixing. Trace caller (wrong data) vs callee (wrong handling). Fix at responsible layer — never patch symptom site.
Grep ALL removed names after extraction/refactoring. Primary file "done" ≠ secondary files clean. Grep entire scope for every removed symbol before declaring complete.
Assume existing values are intentional — ask WHY before changing. Pattern-matching as "wrong" skips context. Before changing any constant/limit/flag: read comments, git blame, surrounding code.
Verify ALL affected outputs, not just the first. One build green ≠ all green. Multi-stack changes (backend/frontend/tests/docs) require verifying EVERY output.
Evaluate fit before copying a nearby pattern. Closest example ≠ matching preconditions — verify the new context shares the same constraints, base classes, scope, lifetime.
Holistic-first debugging — resist nearest-attention trap. Don't dive into first plausible cause. List EVERY precondition (config, env vars, paths, DB, endpoints, creds, versions, DI, data). Verify each against evidence (grep/query — not reasoning). Ask "what would falsify this?" — if nothing, it's not a hypothesis. Most expensive failure: going deeper in "obvious" layer while bug sits in layer never questioned.
Surgical changes — apply the diff test (context-aware). Two modes: (1) Bug fix → every line traces to the bug; no restyling; orphan cleanup only for imports YOUR changes made unused. (2) Review/enhancement → implement improvements AND announce as "Enhancement beyond main request: [what]". Never silently scope-creep. Diff test: "Would this line exist if I wasn't asked to do X?" — if no, delete or announce.
Surface ambiguity before coding — don't pick silently. Multiple valid interpretations → present each with effort: "[Request] could mean (1) [N h], (2) [N h]. Which matters?" List scope/format/volume/constraints assumptions first. If simpler path exists, say so. Never silently pick.
[MANDATORY FIRST ACTION] ALWAYS activate a suitable skill or workflow BEFORE responding. Match task against workflow catalog + skill list; invoke via skill invocation or $start-workflow <workflowId>. NEVER answer or write code before checking. Skip = protocol violation.
Why-Review adversarial mindset — apply when reviewing any plan, decision, or design. Default SKEPTIC not VALIDATOR: steel-man a rejected alternative, invert each stated reason ("what does it sacrifice?"), stress-test top 2-3 assumptions, run pre-mortem ("ships, fails in 3 months — what breaks?"), surface 1-2 alternatives author missed. Section presence ≠ quality; quality = causal reasoning + concrete mitigations + evidence, not "it's better" or "monitor closely".
Front-load report-write in sub-agent prompts for large reviews. Many-file sub-agents hit budget before final write — findings lost. Design prompts so: (1) report-write is first explicit deliverable, (2) append per-file/section (not batched), (3) scope bounded so reads don't exhaust budget. Truncated mid-sentence with no report file → spawn narrower scope, don't retry same prompt.
After context compaction, re-verify all prior phase outcomes before continuing. Summaries describe intent, not environment state (git index, filesystem, processes). On resume, FIRST audit: git status, re-read modified files, verify filesystem. Every "completed" claim is an untested hypothesis until evidence confirms.
OOM/memory: check row count before row size. Triage: (1) Unbounded query — no DB filter for trigger? Push filter to DB; eliminates OOM. (2) Large rows? Projection reduces proportionally. Row reduction > projection in ROI.
Keep domain concepts out of generic/shared/infrastructure layers. Reusable layer (shared library, framework, infra module) must reference NO consumer-specific domain concept — tenant/customer/product IDs, business entities, feature rules. Leak compiles + runs → passes review silently while coupling the "reusable" layer to one consumer. Keep shared type domain-free; push domain fields/logic down into the consumer via subclass/composition. — why: a layer coupled to one consumer's domain is no longer reusable.

Agent Skills: Code Review: Current Or Branch Diff

Install this agent skill to your local

Skill Files

Codex Project-Reference Loading (No Hooks)

Quick Summary

Code Review: Current Or Branch Diff

Review Scope

Review Mindset (NON-NEGOTIABLE)

First Principle — Easy to Change

Core Principles (ENFORCE ALL)

Blast Radius Pre-Analysis (MANDATORY FIRST STEP)

Graph-Assisted Change Review

Review Approach (Report-Driven Multi-Phase — CRITICAL)

Phase 5: Docs-Update Triage (CONDITIONAL)

Readability Checklist (MUST ATTENTION evaluate)

Review Checklist

1. Architecture Compliance (MUST ATTENTION)

2. Code Quality & Clean Code (MUST ATTENTION)

2.5. Naming Conventions (MUST ATTENTION)

3. Project-Specific Patterns (MUST ATTENTION)

4. Security (MUST ATTENTION)

5. Performance (MUST ATTENTION)

6. Common Issues (MUST ATTENTION)

6.5 Bugfix Debugger Trace Gate (MUST ATTENTION)

7. Documentation Staleness (MUST ATTENTION)

8. M1-M6 Compliance Gate — Code-to-Spec Drift (BLOCKING, MUST ATTENTION)

Output Format

Systematic Review Protocol (for 10+ changed files)

Workflow Recommendation

Architecture Boundary Check

Phase 6: Why-Review Findings Validation Gate (MANDATORY before fixing findings)

Phase 7: Recursive Auto-Fix + Full Re-Review Loop (MANDATORY when validated findings remain)

Phase 8: Mandatory Final Docs-Update Gate (MANDATORY — always runs after the review/fix loop converges clean)

Next Steps

AI Agent Integrity Gate (NON-NEGOTIABLE)

Related Skills

Standalone Chain

Subagent Type Selection

Canonical Agent Call Template (Copy Verbatim)

Rules

Closing Reminders

Hookless Prompt Protocol Mirror (Auto-Synced)

[WORKFLOW-EXECUTION-PROTOCOL] [BLOCKING] Workflow Execution Protocol — MANDATORY IMPORTANT MUST CRITICAL. Do not skip for any reason.

Shared AI-SDD Protocol Markers

SYNC:ai-sdd-artifact-contract

SYNC:ai-sdd-artifact-contract:reminder

[LESSON-LEARNED-REMINDER] [BLOCKING] Task Planning & Continuous Improvement — MANDATORY. Do not skip.

Common AI Mistake Prevention (System Lessons)