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:
- Project structure/architecture/tech-stack/deployment/setup (any layer — backend, frontend, or infra):
project-structure-reference.md
- Backend/CQRS/API/domain/entity changes:
backend-patterns-reference.md, domain-entities-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.
Quick Summary
Goal: Deliver a complete, prioritized map of every file relevant to the task via fast, parallel codebase discovery — grep + graph combined — so downstream work starts with full coverage and zero blind spots.
Summary:
- Main step pipeline (run in order, never merge): Phase 0 classify scope → Step 1 analyze request (extract entities/keywords/file-types) → Step 2 spawn parallel
scout subagents (grep/glob) → Step 3 graph-expand 2-3 key files (YOU, main agent) → Step 4 low-result check (<5 files → broader re-pass) → Step 5 synthesize numbered prioritized list → ask the user directly for next step.
- Classify scope FIRST (Phase 0: backend/frontend/full-stack) so you spawn only the agents you need — never the default 3 when the prompt names one layer.
- Sub-agents do the parallel grep/glob; only YOU (main agent) run the graph commands afterward — graph expansion on 2-3 key files is MANDATORY when
.code-graph/graph.db exists and is the step that finds what grep can't.
- This is discovery, not analysis: return prioritized file paths fast (3-5 min), no content deep-dives — that's
investigate's job.
- If <5 files surface, re-examine keywords and run a second pass with broader synonyms before synthesizing; close by asking the user (ask the user directly) which next step to take.
Workflow:
- Phase 0: Classify — Detect search scope (backend/frontend/both) + keyword type
- Analyze Request — Extract entity names, feature keywords, file types from prompt
- Parallel Search — Spawn agents searching backend core, backend infra, frontend paths
- Graph Expand (MANDATORY — DO NOT SKIP) — MUST ATTENTION run graph commands on 2-3 key files. Graph reveals complete dependency network grep CANNOT find.
- Low-Result Check — If <5 files returned, re-examine keywords and run second pass
- Synthesize — Combine grep + graph into numbered, prioritized file list
Key Rules:
- Speed over depth — return file paths only, no content analysis
- Target 3-5 minutes total; 3-minute timeout per agent
- NEVER skip graph expansion when
.code-graph/graph.db exists
--ext / --engine=external switches the search engine to the scout-external agent (gemini/opencode CLIs); default is internal subagents
Scout — Fast Codebase File Discovery
Phase 0: Classify Search Scope
Before spawning agents, classify request:
| Scope |
Detection |
Agent Strategy |
| Backend-only |
server-side class names, domain entities, API handlers |
Agents 1+2, skip Agent 3 |
| Frontend-only |
component names, client-side source, UI features |
Agent 3 only |
| Full-stack |
Feature name spanning both layers |
All 3 agents |
| Unknown |
Ambiguous prompt |
Default all 3 agents |
Think: Prompt mention specific layer? Entity exist in backend, frontend, or both? Adjust agent count — avoid spawning unnecessary agents.
When to Use
- Quickly locating relevant files across large codebase
- Beginning work on features spanning multiple directories
- Before changes affecting multiple parts
- Mapping file landscape before investigation or implementation
NOT for: Deep code analysis (use investigate), debugging (use debug-investigate), implementation (use workflow-feature).
Workflow
Step 1: Analyze Search Request
Extract from USER_PROMPT:
- Entity names (User, Customer, Order)
- Feature names (authentication, notification)
- File types needed (backend, frontend, or both)
Step 2: Execute Parallel Search
Spawn SCALE number of scout subagents in parallel via spawn_agent tool (agent_type: "scout").
WHY scout not Explore: Custom scout agents read .claude/agents/scout.md — includes graph CLI knowledge + Bash access. Built-in Explore agents have NO graph awareness.
Engine Selection (--ext / --engine=external)
Detect the engine flag in args (default: internal):
- Internal (default): spawn
agent_type: "scout" — the parallel grep/glob/graph search described below.
- External (
--ext or --engine=external): spawn agent_type: "scout-external" instead. That agent (.claude/agents/scout-external.md) owns the gemini/opencode CLI dispatch, the Explore fallback, and the install prompt when those CLIs are absent.
Flag only switches which subagent runs. Orchestration identical for both engines: Phase 0 classify, Step 3 graph-expand (run by you, main agent), low-result check, synthesize. Output contract (numbered, prioritized file list) same.
Agent Distribution
- Agent 1 - Backend Core:
{module-source-root}/ domain folder + command folder + query folder (per the project's structure reference)
- Agent 2 - Backend Infra:
{module-source-root}/ event-handler folder + controllers + background-jobs folder (per the project's structure reference)
- Agent 3 - Frontend:
{frontend-apps-dir}/, {frontend-libs-dir}/{domain-lib}/, {frontend-libs-dir}/{common-lib}/
Per agent: 3-minute timeout. Return file paths only — no content analysis. Use Glob (patterns), Grep (content), Bash (graph CLI).
Search-Axis Fan-Out (ONE wave — each axis is blind to the others)
Partitioning by path still searches ONE way. Each axis below finds files the others structurally CANNOT see — that blindness is exactly why a single search angle misses files. Select the axes the prompt warrants (they compose with the path split above: an agent gets one path scope + one axis) and spawn the whole set TOGETHER in one message:
| Axis |
The agent searches by |
Finds what the other axes miss |
| Filename / glob |
Glob on entity + feature names and the project's path conventions |
files NAMED for the feature that never mention the keyword inside |
| Symbol / content grep |
Grep for class, command, query, handler, endpoint, message names |
files that USE the concept without carrying it in their path |
| Convention / naming |
the project's suffix/prefix conventions (*Handler, *Consumer, *-store, per the structure reference) |
siblings that convention says must change together |
| Test coverage |
test directories for the same keywords, plus fixtures/builders/mocks |
the tests pinning the behavior — usually the fastest statement of intent |
| Config / registration |
DI wiring, routes, feature flags, config keys, resource/i18n files |
registrations and string-literal references a symbol grep skips |
Graph is the sixth axis and it stays with YOU — Step 3, after the barrier; sub-agents do not run it here.
Wave rules: precompute the full path × triggered-axis assignment set, then dispatch it in consecutive capacity-bounded waves with one unique exact artifact path per assignment. Never drop an axis to fit host capacity; queue overflow assignments into the next wave. Each wave is read-only with zero write overlap, launches in one message, and reaches its barrier before Step 3. An axis the prompt gives no trigger for is not an assignment and counts as returned. Both engines take identical axis briefs — --ext only swaps scout for scout-external.
Step 3: Graph Expand (MANDATORY — DO NOT SKIP)
YOU (main agent) MUST ATTENTION run graph commands YOURSELF after sub-agents return. NOT optional — without graph, results are incomplete. Sub-agents cannot use graph — only main agent can.
# Check graph exists
ls .code-graph/graph.db 2>/dev/null && echo "GRAPH_AVAILABLE" || echo "NO_GRAPH"
If GRAPH_AVAILABLE, pick 2-3 key files from sub-agent results (entities, commands, bus messages):
# Full dependency network of key file
python .claude/scripts/code_graph connections <key_file> --json
# All callers of key command/handler
python .claude/scripts/code_graph query callers_of <FunctionName> --json
# All importers of bus message class
python .claude/scripts/code_graph query importers_of <file_path> --json
# Batch query multiple files (most efficient)
python .claude/scripts/code_graph batch-query <file1> <file2> <file3> --json
# If graph returns "ambiguous" — disambiguate first
python .claude/scripts/code_graph search <keyword> --kind Function --json
# Trace shortest path between two nodes
python .claude/scripts/code_graph find-path <source_qn> <target_qn> --json
# Filter by service, limit results
python .claude/scripts/code_graph query callers_of <name> --limit 5 --filter "ServiceName" --json
Grep-First Discovery (semantic queries): When prompt describes behavior/flow (not specific file), grep key terms FIRST to discover entry files, then use those as graph input:
- Grep class names, commands, handlers, endpoints
- Use discovered files as input to
connections, batch-query, or trace
- Use
trace --direction both on middle files (controllers, commands) for full upstream + downstream
Graph results get HIGHER priority than grep — structural relationships > text matches. After graph expansion, grep again to verify content in discovered files.
Post-Grep Trace Trigger: whenever a grep/glob surfaces an important entry-point file — an entity, command, query, event/command handler, controller, bus message/consumer, component, store, or api-service — immediately run a graph trace on it before concluding. The trace reveals callers, consumers, bus messages, event chains, and tests that grep CANNOT find: python .claude/scripts/code_graph trace <key-entry-file> --direction both --json. Pattern: grep finds files → graph trace reveals full system flow → grep verifies specific details.
Step 4: Low-Result Check
If total files found <5 after Steps 2-3:
- Re-examine keywords — too specific? Try broader synonyms
- Spawn second scout agent with alternate search terms
- Try
python .claude/scripts/code_graph search <keyword> --json to find nodes by name
Step 5: Synthesize Results
De-duplicate the union FIRST. The axes overlap by design, so the same file legitimately arrives from several agents. Normalize each path (separators, casing) and merge to exactly ONE entry, recording which axes surfaced it — multi-axis hits rank higher (independent confirmation), while a HIGH-priority file surfaced by a single axis is the coverage risk worth one verifying grep before it ships in the list.
Then combine grep + graph into a numbered, prioritized file list (see Results Format).
Search Patterns by Priority
Substitute folder names + file globs from the project's structure reference / docs/project-config.json. {backend-source-glob} / {frontend-source-glob} are the per-stack source extensions.
# HIGH PRIORITY - Core Logic
**/{entity-folder}/**/*{keyword}*.{backend-source-glob}
**/{command-folder}/**/*{keyword}*.{backend-source-glob}
**/{query-folder}/**/*{keyword}*.{backend-source-glob}
**/{event-handler-folder}/**/*{keyword}*.{backend-source-glob}
**/*{keyword}*{component-suffix}.{frontend-source-glob}
**/*{keyword}*{store-suffix}.{frontend-source-glob}
# MEDIUM PRIORITY - Infrastructure
**/{controllers-folder}/**/*{keyword}*.{backend-source-glob}
**/{background-jobs-folder}/**/*{keyword}*.{backend-source-glob}
**/*Consumer*{keyword}*.{backend-source-glob}
**/*{keyword}*{api-service-suffix}.{frontend-source-glob}
# LOW PRIORITY - Supporting
**/*{keyword}*Helper*.{backend-source-glob}
**/*{keyword}*Service*.{backend-source-glob}
**/*{keyword}*{markup-glob}
Graph Intelligence (MANDATORY when graph.db exists)
Results Format
## Scout Results: {USER_PROMPT}
### High Priority - Core Logic
1. `{module-source-root}/{entity-folder}/{Entity}` — domain entity
2. `{module-source-root}/{command-folder}/{Feature}/Save{Entity}Command` — mutating command
...
### Medium Priority - Infrastructure
10. `{module-source-root}/{controllers-folder}/{Entity}Controller` — endpoint
11. `{module-source-root}/{event-handler-folder}/{Feature}/SendNotificationOn{Entity}CreatedEventHandler` — event handler
...
### Low Priority - Supporting
20. `{module-source-root}/{helpers-folder}/{Entity}Helper` — supporting helper
...
### Frontend Files
30. `{frontend-libs-dir}/{domain-lib}/{configured-feature-path}/{feature}-list.component`
...
**Total Files Found:** {count}
**Search Completed In:** {time}
### Suggested Starting Points
1. `{most relevant file}` - {reason}
2. `{second most relevant}` - {reason}
### End-to-Start Trace Candidates
| Role | Candidate files | Why relevant | Evidence |
| ------------------------------ | --------------- | ------------------------------------------------------ | -------------------------------- |
| Observed final output / reader | `{files}` | `{reader, renderer, assertion, query, aggregate, log}` | `{file:line or search evidence}` |
| Storage / projection / cache | `{files}` | `{state consumed by reader}` | `{file:line or search evidence}` |
| Writer / updater | `{files}` | `{writes final state}` | `{file:line or search evidence}` |
| Consumer / handler / job | `{files}` | `{transforms or schedules writes}` | `{file:line or search evidence}` |
| Producer / origin trigger | `{files}` | `{upstream source of input}` | `{file:line or search evidence}` |
**Feeder-path scan:** list every producer/caller/event/job candidate that may write the same final state. Mark unknown paths explicitly instead of hiding them.
### Unresolved Questions
- {any questions that need clarification}
Quality Standards
| Standard |
Expectation |
| Speed |
Complete in 3-5 minutes |
| Accuracy |
Return only relevant files |
| Coverage |
Search all likely directories |
| Efficiency |
Minimize tool calls |
| Structure |
Always use numbered, prioritized lists |
Workflow Recommendation
MANDATORY MUST ATTENTION — NO EXCEPTIONS: If NOT already in workflow, MUST ATTENTION use ask the user directly to ask user:
- Activate
investigation workflow (Recommended) — scout → investigate
- Execute
$scout directly — run this skill standalone
Next Steps
MANDATORY MUST ATTENTION — NO EXCEPTIONS after completing, MUST ATTENTION use ask the user directly to present:
- "$investigate (Recommended)" — Deep-dive into discovered files
- "$plan" — If scouted files sufficient to start planning
- "Skip, continue manually" — user decides
[IMPORTANT] Use task tracking to break ALL work into small tasks BEFORE starting — including tasks for each file read. This prevents context loss from long files. For simple tasks, AI MUST ATTENTION ask user whether to skip.
docs/project-reference/domain-entities-reference.md — Domain entity catalog, relationships, cross-service sync (read when task involves business entities/models)
External Memory: Complex/lengthy work → write findings incrementally to plans/reports/. Prevents context loss.
Evidence Gate: MANDATORY MUST ATTENTION — every claim, finding, recommendation requires file:line proof with confidence % (>80% act, <80% verify first).
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.
Incremental Result Persistence — MANDATORY for all sub-agents or heavy inline steps processing >3 files.
- Before starting: Create report file
plans/reports/{skill}-{date}-{slug}.md
- After each file/section reviewed: Append findings to report immediately — never hold in memory
- Return to main agent: Summary only (per SYNC:subagent-return-contract) with
Full report: path
- Main agent: Reads report file only when resolving specific blockers
Why: Context cutoff mid-execution loses ALL in-memory findings. Each disk write survives compaction. Partial results are better than no results.
Report naming: plans/reports/{skill-name}-{YYMMDD}-{HHmm}-{slug}.md
Sub-Agent Return Contract — When this skill spawns a sub-agent, the sub-agent MUST return ONLY this structure. Main agent reads only this summary — NEVER requests full sub-agent output inline.
## Sub-Agent Result: [skill-name]
Status: ✅ PASS | ⚠️ PARTIAL | ❌ FAIL
Confidence: [0-100]%
### Findings (Critical/High only — max 10 bullets)
- [severity] [file:line] [finding]
### Actions Taken
- [file changed] [what changed]
### Blockers (if any)
- [blocker description]
Full report: plans/reports/[skill-name]-[date]-[slug].md
Main agent reads Full report file ONLY when: (a) resolving a specific blocker, or (b) building a fix plan.
Sub-agent writes full report incrementally (per SYNC:incremental-persistence) — not held in memory.
Context budget — the return payload is a SUMMARY, not a transcript: ≤10 finding bullets, no raw file contents / full diffs / verbatim logs inline, no re-pasted source. Everything beyond the summary lives in the Full report on disk. A sub-agent that would exceed the summary shape MUST write the detail to its report and return only the pointer — the orchestrator's context is the scarce resource the whole map-reduce protects.
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.
- Read
docs/project-config.json first — the project's machine-readable map. It is the single source of truth for THIS repo (modules/paths, framework + search keywords, test/E2E/integration run-commands, design system, architecture rules, workflow patterns); ground exact paths, run-commands, and conventions on it before investigating, planning, or coding — never assume framework defaults (CLAUDE.md + reference docs are derived from it). If it — or the docs index, lessons.md, CLAUDE.md, AGENTS.md, or any required reference doc — is missing or stale, auto-run $project-init or the narrow route ($project-config, $docs-init, $scan-all, $scan --target=<key>, $claude-md-init) first; if Codex mirrors or AGENTS.md are stale, ask the user to run $sync-codex (never auto-run it).
- 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, then before target work state:
Reference docs read: ... | Not applicable: ....
Ready when: scope evaluated, docs/project-config.json consulted, 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.
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.
Evidence-Based Reasoning — Speculation is FORBIDDEN. Every claim needs proof.
- Cite
file:line, grep results, or framework docs for EVERY claim
- Declare confidence: >80% act freely, 60-80% verify first, <60% DO NOT recommend
- Cross-service validation required for architectural changes
- "I don't have enough evidence" is valid and expected output
BLOCKED until: - [ ] Evidence file path (file:line) - [ ] 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: [...]."
Cross-Service Check — Microservices/event-driven: MANDATORY before concluding investigation, plan, spec, or feature doc. Missing downstream consumer = silent regression.
| Boundary |
Grep terms |
| Event producers |
Publish, Dispatch, Send, emit, EventBus, outbox, IntegrationEvent |
| Event consumers |
Consumer, EventHandler, Subscribe, @EventListener, inbox |
| Sagas/orchestration |
Saga, ProcessManager, Choreography, Workflow, Orchestrator |
| Sync service calls |
HTTP/gRPC calls to/from other services |
| Shared contracts |
OpenAPI spec, proto, shared DTO — flag breaking changes |
| Data ownership |
Other service reads/writes same table/collection → Shared-DB anti-pattern |
Per touchpoint: owner service · message name · consumers · risk (NONE / ADDITIVE / BREAKING).
BLOCKED until: Producers scanned · Consumers scanned · Sagas checked · Contracts reviewed · Breaking-change risk flagged
Rationalization Prevention — AI skips steps via these evasions. Recognize and reject:
| Evasion |
Rebuttal |
| "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. |
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 OR flagging one as a defect. Before changing or reporting a constant, limit, flag, cutoff, wording, or pattern, read nearby context and history, the CALLER's ordering, and 2+ sibling call sites of the same convention. A doc stating WHAT without WHY is missing rationale, not proof of a missing guard.
Surface ambiguity before acting — don't pick silently. Multiple valid interpretations require an explicit question or stated assumption with risk.
Assert the outcome your system owns, not the intermediate state your infrastructure owns. When verifying async work, assert the final business state — never the delivery/retry bookkeeping held in shared infrastructure that any co-running process can write. Such a check passes when run alone and flakes the moment anything else shares that infrastructure.
Keep shared guidance role-relevant. Universal guidance must help every receiving skill or agent; code-specific obligations belong only in code-specific protocols.
MUST ATTENTION cite file:line evidence for every claim. Confidence >80% to act, <60% = do NOT recommend.
MUST ATTENTION never skip steps via evasions. Plan anyway. Test first. Show grep evidence with file:line.
MUST ATTENTION run at least ONE graph command on key files before concluding when .code-graph/graph.db exists.
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 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 Before investigating, planning, or coding, read
docs/project-config.json (the project map: modules/paths, run-commands, conventions, architecture/workflow rules) + the required project-reference docs, and cite Reference docs read: ....
- MANDATORY Always include
lessons.md; project config + conventions override generic framework 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.
- 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.
Prompt-Enhance Closing Anchors
IMPORTANT MUST ATTENTION follow declared step order for this skill; NEVER skip, reorder, or merge steps without explicit user approval
IMPORTANT MUST ATTENTION for every step/sub-skill call: set in_progress before execution, set completed after execution
IMPORTANT MUST ATTENTION every skipped step MUST include explicit reason; every completed step MUST include concise evidence
IMPORTANT MUST ATTENTION if Task tools unavailable, maintain an equivalent step-by-step plan tracker with synchronized statuses
Parallel Sub-Agent Dispatch — Plan parallelism the moment a task breakdown exists, BEFORE executing it — running provably independent tasks sequentially wastes wall-clock. Applies to every multi-step job: workflow steps, planning, batch updates, investigation, research, scans, reviews, doc sync. Plan execution is metadata-gated, NEVER default-parallel — fan-out follows ONLY what the plan declares (PAR/SEQ tags + per-phase write set); an untagged plan runs sequentially — why: a derived write set cannot see cascade or generated writes.
- Tag every task
PAR or SEQ. PAR = inputs exclude every pending task's output AND write set disjoint from every other PAR. Else SEQ — MUST ATTENTION name the dependency forcing it.
- Group
PAR into waves. No edge between members. Two writers of one file NEVER share a wave. Read-only work (search, investigation, review, research) parallelizes freely.
- Declare before dispatch:
Parallel plan: wave 1 = [...] · wave 2 = [...] · SEQ = [...] (reason).
- Spawn each wave in ONE message — every
spawn_agent call in one response, NEVER dripped per turn. Route each task to its specialist (.claude/skills/shared/sub-agent-selection-guide.md); NEVER code-reviewer as catch-all.
- Brief each sub-agent self-contained: goal · scope + owned files · reference docs · return contract (summary +
Full report: path, per SYNC:subagent-return-contract) · incremental persistence to plans/reports/ (per SYNC:incremental-persistence).
- Barrier per wave. Advance ONLY after EVERY member returns (a skipped conditional counts as returned). Merge, mark each task completed/skipped, THEN dispatch the next wave. Mutating steps wait for the barrier.
- One level deep. A dispatched sub-agent executes its own brief; further fan-out stays the orchestrator's job unless that agent's
.claude/agents/*.md definition authorizes it.
NEVER parallelize: tasks sharing a write target · a task consuming a pending task's output · trivial single-file work (dispatch overhead > gain) · an order a skill or workflow explicitly fixes · gates awaiting user approval.
Blocked until: MUST ATTENTION every task tagged PAR/SEQ with a named reason per SEQ · waves declared + write-set disjointness checked · each wave spawned in ONE message · barrier honored before the next wave.
- MANDATORY After planning tasks, tag each PAR/SEQ and spawn every PAR wave as parallel sub-agents in ONE message — default parallel for workflows, batch updates, investigation, research, reviews; plan execution fans out ONLY on what the plan declares.
- MANDATORY Disjoint write sets per wave · all-return barrier before the next wave · specialist routing · sub-agents NEVER fan out further unless their own agent definition authorizes it.
Project Protocol Overlay — Before executing this skill, resolve any PROJECT overlay rules layered onto it: match this sk
…(truncated)
1---
2name: scout
3description: '[Investigation] Use when quickly locating relevant files and affected areas across a large codebase.'
4metadata:
5 source: https://skillmd.com/skills/duc01226/scout
6---
7
8> Codex compatibility note:
9>
10> - Invoke repository skills with `$skill-name` in Codex; this mirrored copy rewrites legacy Claude `/skill-name` references.
11> - Task tracker mandate: BEFORE executing any workflow or skill step, create/update task tracking for all steps and keep it synchronized as progress changes.
12> - User-question prompts mean to ask the user directly in Codex.
13> - Ignore Claude-specific mode-switch instructions when they appear.
14> - Strict execution contract: when a user explicitly invokes a skill, execute that skill protocol as written.
15> - 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.
16> - Do not skip, reorder, or merge protocol steps unless the user explicitly approves the deviation first.
17> - For workflow skills, execute each listed child-skill step explicitly and report step-by-step evidence.
18> - If a required step/tool cannot run in this environment, stop and ask the user before adapting.
19
20<!-- CODEX:PROJECT-REFERENCE-LOADING:START -->
21
22## Codex Project-Reference Loading (No Hooks)
23
24Codex uses static project-reference loading instead of runtime-injected project docs.
25When coding, planning, debugging, testing, or reviewing, open project docs explicitly using this routing.
26
27**Always read:**
28
29- `docs/project-config.json` (project-specific paths, commands, modules, and workflow/test settings)
30- `docs/project-reference/docs-index-reference.md` (routes to the full `docs/project-reference/*` catalog)
31- `docs/project-reference/lessons.md` (always-on guardrails and anti-patterns)
32
33**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.
34
35**Situation-based docs:**
36
37- Project structure/architecture/tech-stack/deployment/setup (any layer — backend, frontend, or infra): `project-structure-reference.md`
38- Backend/CQRS/API/domain/entity changes: `backend-patterns-reference.md`, `domain-entities-reference.md`
39- Frontend/UI/styling/design-system: `frontend-patterns-reference.md`, `scss-styling-guide.md`, `design-system/README.md`
40- Spec authoring, `docs/specs/` pathing, or TC format: `feature-spec-reference.md`, `spec-system-reference.md`, `spec-principles.md`
41- Behavior/public-contract changes or spec-test-code sync: `workflow-spec-test-code-cycle-reference.md` plus the spec docs above
42- Derived spec indexes/ERDs/reimplementation guides: `spec-system-reference.md` and source Feature Specs under `docs/specs/`
43- Integration test implementation/review: `integration-test-reference.md`
44- E2E test implementation/review: `e2e-test-reference.md`
45- Code review/audit work: `code-review-rules.md` plus domain docs above based on changed files
46
47Do not read all docs blindly. Start from `docs-index-reference.md`, then open only relevant files for the task.
48
49<!-- CODEX:PROJECT-REFERENCE-LOADING:END -->
50
51<!-- PROMPT-ENHANCE:STEP-TASK-ANCHOR:START -->
52
53> **[BLOCKING]** Execute skill steps in declared order. NEVER skip, reorder, or merge steps without explicit user approval.
54> **[BLOCKING]** Before each step or sub-skill call, update task tracking: set `in_progress` when step starts, set `completed` when step ends.
55> **[BLOCKING]** Every completed/skipped step MUST include brief evidence or explicit skip reason.
56> **[BLOCKING]** If Task tools are unavailable, create and maintain an equivalent step-by-step plan tracker with the same status transitions.
57
58<!-- PROMPT-ENHANCE:STEP-TASK-ANCHOR:END -->
59
60## Quick Summary
61
62**Goal:** Deliver a complete, prioritized map of every file relevant to the task via fast, parallel codebase discovery — grep + graph combined — so downstream work starts with full coverage and zero blind spots.
63
64**Summary:**
65
66- **Main step pipeline (run in order, never merge):** Phase 0 classify scope → Step 1 analyze request (extract entities/keywords/file-types) → Step 2 spawn parallel `scout` subagents (grep/glob) → Step 3 graph-expand 2-3 key files (YOU, main agent) → Step 4 low-result check (<5 files → broader re-pass) → Step 5 synthesize numbered prioritized list → ask the user directly for next step.
67- Classify scope FIRST (Phase 0: backend/frontend/full-stack) so you spawn only the agents you need — never the default 3 when the prompt names one layer.
68- Sub-agents do the parallel grep/glob; only YOU (main agent) run the graph commands afterward — graph expansion on 2-3 key files is MANDATORY when `.code-graph/graph.db` exists and is the step that finds what grep can't.
69- This is discovery, not analysis: return prioritized file paths fast (3-5 min), no content deep-dives — that's `investigate`'s job.
70- If <5 files surface, re-examine keywords and run a second pass with broader synonyms before synthesizing; close by asking the user (ask the user directly) which next step to take.
71
72**Workflow:**
73
741. **Phase 0: Classify** — Detect search scope (backend/frontend/both) + keyword type
752. **Analyze Request** — Extract entity names, feature keywords, file types from prompt
763. **Parallel Search** — Spawn agents searching backend core, backend infra, frontend paths
774. **Graph Expand (MANDATORY — DO NOT SKIP)** — **MUST ATTENTION** run graph commands on 2-3 key files. Graph reveals complete dependency network grep CANNOT find.
785. **Low-Result Check** — If <5 files returned, re-examine keywords and run second pass
796. **Synthesize** — Combine grep + graph into numbered, prioritized file list
80
81**Key Rules:**
82
83- Speed over depth — return file paths only, no content analysis
84- Target 3-5 minutes total; 3-minute timeout per agent
85- NEVER skip graph expansion when `.code-graph/graph.db` exists
86- `--ext` / `--engine=external` switches the search engine to the `scout-external` agent (gemini/opencode CLIs); default is internal subagents
87
88# Scout — Fast Codebase File Discovery
89
90---
91
92## Phase 0: Classify Search Scope
93
94**Before spawning agents**, classify request:
95
96| Scope | Detection | Agent Strategy |
97| ------------- | ------------------------------------------------------ | ------------------------ |
98| Backend-only | server-side class names, domain entities, API handlers | Agents 1+2, skip Agent 3 |
99| Frontend-only | component names, client-side source, UI features | Agent 3 only |
100| Full-stack | Feature name spanning both layers | All 3 agents |
101| Unknown | Ambiguous prompt | Default all 3 agents |
102
103**Think:** Prompt mention specific layer? Entity exist in backend, frontend, or both? Adjust agent count — avoid spawning unnecessary agents.
104
105---
106
107## When to Use
108
109- Quickly locating relevant files across large codebase
110- Beginning work on features spanning multiple directories
111- Before changes affecting multiple parts
112- Mapping file landscape before investigation or implementation
113
114**NOT for:** Deep code analysis (use `investigate`), debugging (use `debug-investigate`), implementation (use `workflow-feature`).
115
116---
117
118## Workflow
119
120### Step 1: Analyze Search Request
121
122Extract from USER_PROMPT:
123
124- Entity names (User, Customer, Order)
125- Feature names (authentication, notification)
126- File types needed (backend, frontend, or both)
127
128### Step 2: Execute Parallel Search
129
130Spawn SCALE number of `scout` subagents in parallel via `spawn_agent` tool (`agent_type: "scout"`).
131
132**WHY `scout` not `Explore`:** Custom `scout` agents read `.claude/agents/scout.md` — includes graph CLI knowledge + Bash access. Built-in `Explore` agents have NO graph awareness.
133
134#### Engine Selection (`--ext` / `--engine=external`)
135
136Detect the engine flag in args (default: **internal**):
137
138- **Internal (default):** spawn `agent_type: "scout"` — the parallel grep/glob/graph search described below.
139- **External (`--ext` or `--engine=external`):** spawn `agent_type: "scout-external"` instead. That agent (`.claude/agents/scout-external.md`) owns the `gemini`/`opencode` CLI dispatch, the Explore fallback, and the install prompt when those CLIs are absent.
140
141Flag only switches **which subagent runs**. Orchestration identical for both engines: Phase 0 classify, Step 3 graph-expand (run by you, main agent), low-result check, synthesize. Output contract (numbered, prioritized file list) same.
142
143#### Agent Distribution
144
145- **Agent 1 - Backend Core**: `{module-source-root}/` domain folder + command folder + query folder (per the project's structure reference)
146- **Agent 2 - Backend Infra**: `{module-source-root}/` event-handler folder + controllers + background-jobs folder (per the project's structure reference)
147- **Agent 3 - Frontend**: `{frontend-apps-dir}/`, `{frontend-libs-dir}/{domain-lib}/`, `{frontend-libs-dir}/{common-lib}/`
148
149Per agent: 3-minute timeout. Return file paths only — no content analysis. Use Glob (patterns), Grep (content), Bash (graph CLI).
150
151#### Search-Axis Fan-Out (ONE wave — each axis is blind to the others)
152
153Partitioning by path still searches ONE way. Each axis below finds files the others structurally CANNOT see — that blindness is exactly why a single search angle misses files. Select the axes the prompt warrants (they compose with the path split above: an agent gets one path scope + one axis) and spawn the whole set TOGETHER in one message:
154
155| Axis | The agent searches by | Finds what the other axes miss |
156| ------------------------- | --------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------ |
157| **Filename / glob** | Glob on entity + feature names and the project's path conventions | files NAMED for the feature that never mention the keyword inside |
158| **Symbol / content grep** | Grep for class, command, query, handler, endpoint, message names | files that USE the concept without carrying it in their path |
159| **Convention / naming** | the project's suffix/prefix conventions (`*Handler`, `*Consumer`, `*-store`, per the structure reference) | siblings that convention says must change together |
160| **Test coverage** | test directories for the same keywords, plus fixtures/builders/mocks | the tests pinning the behavior — usually the fastest statement of intent |
161| **Config / registration** | DI wiring, routes, feature flags, config keys, resource/i18n files | registrations and string-literal references a symbol grep skips |
162
163**Graph is the sixth axis and it stays with YOU** — Step 3, after the barrier; sub-agents do not run it here.
164
165Wave rules: precompute the full path × triggered-axis assignment set, then dispatch it in consecutive **capacity-bounded waves** with one unique exact artifact path per assignment. Never drop an axis to fit host capacity; queue overflow assignments into the next wave. Each wave is read-only with zero write overlap, launches in one message, and reaches its barrier before Step 3. An axis the prompt gives no trigger for is not an assignment and counts as returned. Both engines take identical axis briefs — `--ext` only swaps `scout` for `scout-external`.
166
167### Step 3: Graph Expand (MANDATORY — DO NOT SKIP)
168
169**YOU (main agent) MUST ATTENTION run graph commands YOURSELF after sub-agents return.** NOT optional — without graph, results are incomplete. Sub-agents cannot use graph — only main agent can.
170
171```bash
172# Check graph exists
173ls .code-graph/graph.db 2>/dev/null && echo "GRAPH_AVAILABLE" || echo "NO_GRAPH"
174```
175
176If GRAPH_AVAILABLE, pick 2-3 key files from sub-agent results (entities, commands, bus messages):
177
178```bash
179# Full dependency network of key file
180python .claude/scripts/code_graph connections <key_file> --json
181
182# All callers of key command/handler
183python .claude/scripts/code_graph query callers_of <FunctionName> --json
184
185# All importers of bus message class
186python .claude/scripts/code_graph query importers_of <file_path> --json
187
188# Batch query multiple files (most efficient)
189python .claude/scripts/code_graph batch-query <file1> <file2> <file3> --json
190
191# If graph returns "ambiguous" — disambiguate first
192python .claude/scripts/code_graph search <keyword> --kind Function --json
193
194# Trace shortest path between two nodes
195python .claude/scripts/code_graph find-path <source_qn> <target_qn> --json
196
197# Filter by service, limit results
198python .claude/scripts/code_graph query callers_of <name> --limit 5 --filter "ServiceName" --json
199```
200
201**Grep-First Discovery (semantic queries):** When prompt describes behavior/flow (not specific file), grep key terms FIRST to discover entry files, then use those as graph input:
202
2031. Grep class names, commands, handlers, endpoints
2042. Use discovered files as input to `connections`, `batch-query`, or `trace`
2053. Use `trace --direction both` on middle files (controllers, commands) for full upstream + downstream
206
207Graph results get HIGHER priority than grep — structural relationships > text matches. After graph expansion, grep again to verify content in discovered files.
208
209**Post-Grep Trace Trigger:** whenever a grep/glob surfaces an important entry-point file — an entity, command, query, event/command handler, controller, bus message/consumer, component, store, or api-service — immediately run a graph trace on it before concluding. The trace reveals callers, consumers, bus messages, event chains, and tests that grep CANNOT find: `python .claude/scripts/code_graph trace <key-entry-file> --direction both --json`. **Pattern: grep finds files → graph trace reveals full system flow → grep verifies specific details.**
210
211### Step 4: Low-Result Check
212
213If total files found <5 after Steps 2-3:
214
2151. Re-examine keywords — too specific? Try broader synonyms
2162. Spawn second scout agent with alternate search terms
2173. Try `python .claude/scripts/code_graph search <keyword> --json` to find nodes by name
218
219### Step 5: Synthesize Results
220
221**De-duplicate the union FIRST.** The axes overlap by design, so the same file legitimately arrives from several agents. Normalize each path (separators, casing) and merge to exactly ONE entry, recording which axes surfaced it — multi-axis hits rank higher (independent confirmation), while a HIGH-priority file surfaced by a single axis is the coverage risk worth one verifying grep before it ships in the list.
222
223Then combine grep + graph into a numbered, prioritized file list (see Results Format).
224
225---
226
227## Search Patterns by Priority
228
229> Substitute folder names + file globs from the project's structure reference / `docs/project-config.json`. `{backend-source-glob}` / `{frontend-source-glob}` are the per-stack source extensions.
230
231```
232# HIGH PRIORITY - Core Logic
233**/{entity-folder}/**/*{keyword}*.{backend-source-glob}
234**/{command-folder}/**/*{keyword}*.{backend-source-glob}
235**/{query-folder}/**/*{keyword}*.{backend-source-glob}
236**/{event-handler-folder}/**/*{keyword}*.{backend-source-glob}
237**/*{keyword}*{component-suffix}.{frontend-source-glob}
238**/*{keyword}*{store-suffix}.{frontend-source-glob}
239
240# MEDIUM PRIORITY - Infrastructure
241**/{controllers-folder}/**/*{keyword}*.{backend-source-glob}
242**/{background-jobs-folder}/**/*{keyword}*.{backend-source-glob}
243**/*Consumer*{keyword}*.{backend-source-glob}
244**/*{keyword}*{api-service-suffix}.{frontend-source-glob}
245
246# LOW PRIORITY - Supporting
247**/*{keyword}*Helper*.{backend-source-glob}
248**/*{keyword}*Service*.{backend-source-glob}
249**/*{keyword}*{markup-glob}
250```
251
252---
253
254## Graph Intelligence (MANDATORY when graph.db exists)
255
256---
257
258## Results Format
259
260```markdown
261## Scout Results: {USER_PROMPT}
262
263### High Priority - Core Logic
264
2651. `{module-source-root}/{entity-folder}/{Entity}` — domain entity
2662. `{module-source-root}/{command-folder}/{Feature}/Save{Entity}Command` — mutating command
267 ...
268
269### Medium Priority - Infrastructure
270
27110. `{module-source-root}/{controllers-folder}/{Entity}Controller` — endpoint
27211. `{module-source-root}/{event-handler-folder}/{Feature}/SendNotificationOn{Entity}CreatedEventHandler` — event handler
273 ...
274
275### Low Priority - Supporting
276
27720. `{module-source-root}/{helpers-folder}/{Entity}Helper` — supporting helper
278 ...
279
280### Frontend Files
281
28230. `{frontend-libs-dir}/{domain-lib}/{configured-feature-path}/{feature}-list.component`
283 ...
284
285**Total Files Found:** {count}
286**Search Completed In:** {time}
287
288### Suggested Starting Points
289
2901. `{most relevant file}` - {reason}
2912. `{second most relevant}` - {reason}
292
293### End-to-Start Trace Candidates
294
295| Role | Candidate files | Why relevant | Evidence |
296| ------------------------------ | --------------- | ------------------------------------------------------ | -------------------------------- |
297| Observed final output / reader | `{files}` | `{reader, renderer, assertion, query, aggregate, log}` | `{file:line or search evidence}` |
298| Storage / projection / cache | `{files}` | `{state consumed by reader}` | `{file:line or search evidence}` |
299| Writer / updater | `{files}` | `{writes final state}` | `{file:line or search evidence}` |
300| Consumer / handler / job | `{files}` | `{transforms or schedules writes}` | `{file:line or search evidence}` |
301| Producer / origin trigger | `{files}` | `{upstream source of input}` | `{file:line or search evidence}` |
302
303**Feeder-path scan:** list every producer/caller/event/job candidate that may write the same final state. Mark unknown paths explicitly instead of hiding them.
304
305### Unresolved Questions
306
307- {any questions that need clarification}
308```
309
310---
311
312## Quality Standards
313
314| Standard | Expectation |
315| -------------- | -------------------------------------- |
316| **Speed** | Complete in 3-5 minutes |
317| **Accuracy** | Return only relevant files |
318| **Coverage** | Search all likely directories |
319| **Efficiency** | Minimize tool calls |
320| **Structure** | Always use numbered, prioritized lists |
321
322---
323
324## Workflow Recommendation
325
326> **MANDATORY MUST ATTENTION — NO EXCEPTIONS:** If NOT already in workflow, MUST ATTENTION use ask the user directly to ask user:
327>
328> 1. **Activate `investigation` workflow** (Recommended) — scout → investigate
329> 2. **Execute `$scout` directly** — run this skill standalone
330
331---
332
333## Next Steps
334
335**MANDATORY MUST ATTENTION — NO EXCEPTIONS** after completing, MUST ATTENTION use ask the user directly to present:
336
337- **"$investigate (Recommended)"** — Deep-dive into discovered files
338- **"$plan"** — If scouted files sufficient to start planning
339- **"Skip, continue manually"** — user decides
340
341---
342
343> **[IMPORTANT]** Use task tracking to break ALL work into small tasks BEFORE starting — including tasks for each file read. This prevents context loss from long files. For simple tasks, AI MUST ATTENTION ask user whether to skip.
344
345- `docs/project-reference/domain-entities-reference.md` — Domain entity catalog, relationships, cross-service sync (read when task involves business entities/models)
346
347> **External Memory:** Complex/lengthy work → write findings incrementally to `plans/reports/`. Prevents context loss.
348
349> **Evidence Gate:** MANDATORY MUST ATTENTION — every claim, finding, recommendation requires `file:line` proof with confidence % (>80% act, <80% verify first).
350
351<!-- SYNC:graph-assisted-investigation -->
352
353> **Graph-Assisted Investigation** — MANDATORY when `.code-graph/graph.db` exists.
354>
355> **HARD-GATE:** MUST ATTENTION run at least ONE graph command on key files before concluding any investigation.
356>
357> **Pattern:** Grep finds files → `trace --direction both` reveals full system flow → Grep verifies details
358>
359> | Task | Minimum Graph Action |
360> | ------------------- | -------------------------------------------- |
361> | Investigation/Scout | `trace --direction both` on 2-3 entry files |
362> | Fix/Debug | `callers_of` on buggy function + `tests_for` |
363> | Feature/Enhancement | `connections` on files to be modified |
364> | Code Review | `tests_for` on changed functions |
365> | Blast Radius | `trace --direction downstream` |
366>
367> **CLI:** `python .claude/scripts/code_graph {command} --json`. Use `--node-mode file` first (10-30x less noise), then `--node-mode function` for detail.
368
369<!-- /SYNC:graph-assisted-investigation -->
370
371<!-- SYNC:incremental-persistence -->
372
373> **Incremental Result Persistence** — MANDATORY for all sub-agents or heavy inline steps processing >3 files.
374>
375> 1. **Before starting:** Create report file `plans/reports/{skill}-{date}-{slug}.md`
376> 2. **After each file/section reviewed:** Append findings to report immediately — never hold in memory
377> 3. **Return to main agent:** Summary only (per SYNC:subagent-return-contract) with `Full report:` path
378> 4. **Main agent:** Reads report file only when resolving specific blockers
379>
380> **Why:** Context cutoff mid-execution loses ALL in-memory findings. Each disk write survives compaction. Partial results are better than no results.
381>
382> **Report naming:** `plans/reports/{skill-name}-{YYMMDD}-{HHmm}-{slug}.md`
383
384<!-- /SYNC:incremental-persistence -->
385
386<!-- SYNC:subagent-return-contract -->
387
388> **Sub-Agent Return Contract** — When this skill spawns a sub-agent, the sub-agent MUST return ONLY this structure. Main agent reads only this summary — NEVER requests full sub-agent output inline.
389>
390> ```markdown
391> ## Sub-Agent Result: [skill-name]
392>
393> Status: ✅ PASS | ⚠️ PARTIAL | ❌ FAIL
394> Confidence: [0-100]%
395>
396> ### Findings (Critical/High only — max 10 bullets)
397>
398> - [severity] [file:line] [finding]
399>
400> ### Actions Taken
401>
402> - [file changed] [what changed]
403>
404> ### Blockers (if any)
405>
406> - [blocker description]
407>
408> Full report: plans/reports/[skill-name]-[date]-[slug].md
409> ```
410>
411> Main agent reads `Full report` file ONLY when: (a) resolving a specific blocker, or (b) building a fix plan.
412> Sub-agent writes full report incrementally (per SYNC:incremental-persistence) — not held in memory.
413>
414> **Context budget** — the return payload is a SUMMARY, not a transcript: ≤10 finding bullets, no raw file contents / full diffs / verbatim logs inline, no re-pasted source. Everything beyond the summary lives in the `Full report` on disk. A sub-agent that would exceed the summary shape MUST write the detail to its report and return only the pointer — the orchestrator's context is the scarce resource the whole map-reduce protects.
415
416<!-- /SYNC:subagent-return-contract -->
417
418<!-- SYNC:nested-task-creation -->
419
420> **Nested Task Expansion Contract** — For workflow-step invocation, the `[Workflow] ...` row is only a parent container; the child skill still creates visible phase tasks.
421>
422> 1. Call the current task list first. If a matching active parent workflow row exists, set `nested=true` and record `parentTaskId`; otherwise run standalone.
423> 2. Create one task per declared phase before phase work. When nested, prefix subjects `[N.M] $skill-name — phase`.
424> 3. When nested, link the parent with `TaskUpdate(parentTaskId, addBlockedBy: [childIds])`.
425> 4. Orchestrators must pre-expand a child skill's phase list and link the workflow row before invoking that child skill or sub-agent.
426> 5. Mark exactly one child `in_progress` before work and `completed` immediately after evidence is written.
427> 6. Complete the parent only after all child tasks are completed or explicitly cancelled with reason.
428>
429> **Blocked until:** the current task list done, child phases created, parent linked when nested, first child marked `in_progress`.
430
431<!-- /SYNC:nested-task-creation -->
432
433<!-- SYNC:project-reference-docs-guide -->
434
435> **Project Reference Docs Gate** — Run after task-tracking bootstrap and before target/source file reads, grep, edits, or analysis. Project docs override generic framework assumptions.
436>
437> 1. Identify scope: file types, domain area, and operation.
438> 2. **Read `docs/project-config.json` first — the project's machine-readable map.** It is the single source of truth for THIS repo (modules/paths, framework + search keywords, test/E2E/integration run-commands, design system, architecture rules, workflow patterns); ground exact paths, run-commands, and conventions on it **before investigating, planning, or coding** — never assume framework defaults (`CLAUDE.md` + reference docs are derived from it). If it — or the docs index, `lessons.md`, `CLAUDE.md`, `AGENTS.md`, or any required reference doc — is missing or stale, auto-run `$project-init` or the narrow route (`$project-config`, `$docs-init`, `$scan-all`, `$scan --target=<key>`, `$claude-md-init`) first; if Codex mirrors or `AGENTS.md` are stale, ask the user to run `$sync-codex` (never auto-run it).
439> 3. 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`.
440> 4. Read every required doc, then before target work state: `Reference docs read: ... | Not applicable: ...`.
441>
442> **Ready when:** scope evaluated, `docs/project-config.json` consulted, required docs checked/read or setup route completed, `lessons.md` confirmed, citation emitted.
443
444<!-- /SYNC:project-reference-docs-guide -->
445
446<!-- SYNC:task-tracking-external-report -->
447
448> **Task Tracking & External Report Persistence** — Bootstrap this before execution; then run project-reference doc prefetch before target/source work.
449>
450> 1. Create a small task breakdown before target file reads, grep, edits, or analysis. On context loss, inspect the current task list first.
451> 2. Mark one task `in_progress` before work and `completed` immediately after evidence; never batch transitions.
452> 3. For plan/review work, create `plans/reports/{skill}-{YYMMDD}-{HHmm}-{slug}.md` before first finding.
453> 4. Append findings after each file/section/decision and synthesize from the report file at the end.
454> 5. Final output cites `Full report: plans/reports/{filename}`.
455>
456> **Blocked until:** task breakdown exists, report path declared for plan/review work, first finding persisted before the next finding.
457
458<!-- /SYNC:task-tracking-external-report -->
459
460<!-- SYNC:critical-thinking-mindset -->
461
462> **Critical Thinking Mindset** — Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence >80% to act.
463> **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.
464
465<!-- /SYNC:critical-thinking-mindset -->
466
467<!-- SYNC:evidence-based-reasoning -->
468
469> **Evidence-Based Reasoning** — Speculation is FORBIDDEN. Every claim needs proof.
470>
471> 1. Cite `file:line`, grep results, or framework docs for EVERY claim
472> 2. Declare confidence: >80% act freely, 60-80% verify first, <60% DO NOT recommend
473> 3. Cross-service validation required for architectural changes
474> 4. "I don't have enough evidence" is valid and expected output
475>
476> **BLOCKED until:** `- [ ]` Evidence file path (`file:line`) `- [ ]` Grep search performed `- [ ]` 3+ similar patterns found `- [ ]` Confidence level stated
477>
478> **Forbidden without proof:** "obviously", "I think", "should be", "probably", "this is because"
479> **If incomplete →** output: `"Insufficient evidence. Verified: [...]. Not verified: [...]."`
480
481<!-- /SYNC:evidence-based-reasoning -->
482
483<!-- SYNC:cross-service-check -->
484
485> **Cross-Service Check** — Microservices/event-driven: MANDATORY before concluding investigation, plan, spec, or feature doc. Missing downstream consumer = silent regression.
486>
487> | Boundary | Grep terms |
488> | ------------------- | ------------------------------------------------------------------------------- |
489> | Event producers | `Publish`, `Dispatch`, `Send`, `emit`, `EventBus`, `outbox`, `IntegrationEvent` |
490> | Event consumers | `Consumer`, `EventHandler`, `Subscribe`, `@EventListener`, `inbox` |
491> | Sagas/orchestration | `Saga`, `ProcessManager`, `Choreography`, `Workflow`, `Orchestrator` |
492> | Sync service calls | HTTP/gRPC calls to/from other services |
493> | Shared contracts | OpenAPI spec, proto, shared DTO — flag breaking changes |
494> | Data ownership | Other service reads/writes same table/collection → Shared-DB anti-pattern |
495>
496> **Per touchpoint:** owner service · message name · consumers · risk (NONE / ADDITIVE / BREAKING).
497>
498> **BLOCKED until:** Producers scanned · Consumers scanned · Sagas checked · Contracts reviewed · Breaking-change risk flagged
499
500<!-- /SYNC:cross-service-check -->
501
502<!-- SYNC:rationalization-prevention -->
503
504> **Rationalization Prevention** — AI skips steps via these evasions. Recognize and reject:
505>
506> | Evasion | Rebuttal |
507> | ---------------------------- | ------------------------------------------------------------- |
508> | "Too simple for a plan" | Simple + wrong assumptions = wasted time. Plan anyway. |
509> | "I'll test after" | RED before GREEN. Write/verify test first. |
510> | "Already searched" | Show grep evidence with `file:line`. No proof = no search. |
511> | "Just do it" | Still need task tracking. Skip depth, never skip tracking. |
512> | "Just a small fix" | Small fix in wrong location cascades. Verify file:line first. |
513> | "Code is self-explanatory" | Future readers need evidence trail. Document anyway. |
514> | "Combine steps to save time" | Combined steps dilute focus. Each step has distinct purpose. |
515
516<!-- /SYNC:rationalization-prevention -->
517
518<!-- SYNC:ai-mistake-prevention -->
519
520> **AI Mistake Prevention** — Failure modes to avoid on every task:
521>
522> **Re-read files after context changes.** Context compaction, resume, or long-running work can make memory stale; verify current files before acting.
523> **Verify generated content against source evidence.** AI hallucinates APIs, names, claims, and document facts. Check the relevant source before documenting or referencing.
524> **Check downstream references before deleting or renaming.** Removing an artifact can stale docs, generated mirrors, configs, and callers; map references first.
525> **Trace the full impact chain after edits.** Changing a definition can miss derived outputs and consumers. Follow the affected chain before declaring done.
526> **Verify ALL affected outputs, not just the first.** One green check is not all green checks; validate every output surface the change can affect.
527> **Assume existing values are intentional — ask WHY before changing OR flagging one as a defect.** Before changing or reporting a constant, limit, flag, cutoff, wording, or pattern, read nearby context and history, the CALLER's ordering, and 2+ sibling call sites of the same convention. A doc stating WHAT without WHY is missing rationale, not proof of a missing guard.
528> **Surface ambiguity before acting — don't pick silently.** Multiple valid interpretations require an explicit question or stated assumption with risk.
529> **Assert the outcome your system owns, not the intermediate state your infrastructure owns.** When verifying async work, assert the final business state — never the delivery/retry bookkeeping held in shared infrastructure that any co-running process can write. Such a check passes when run alone and flakes the moment anything else shares that infrastructure.
530> **Keep shared guidance role-relevant.** Universal guidance must help every receiving skill or agent; code-specific obligations belong only in code-specific protocols.
531
532<!-- /SYNC:ai-mistake-prevention -->
533
534<!-- SYNC:evidence-based-reasoning:reminder -->
535
536**MUST ATTENTION** cite `file:line` evidence for every claim. Confidence >80% to act, <60% = do NOT recommend.
537
538<!-- /SYNC:evidence-based-reasoning:reminder -->
539
540<!-- SYNC:rationalization-prevention:reminder -->
541
542**MUST ATTENTION** never skip steps via evasions. Plan anyway. Test first. Show grep evidence with `file:line`.
543
544<!-- /SYNC:rationalization-prevention:reminder -->
545
546<!-- SYNC:graph-assisted-investigation:reminder -->
547
548**MUST ATTENTION** run at least ONE graph command on key files before concluding when `.code-graph/graph.db` exists.
549
550<!-- /SYNC:graph-assisted-investigation:reminder -->
551
552<!-- SYNC:critical-thinking-mindset:reminder -->
553
554**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.
555
556<!-- /SYNC:critical-thinking-mindset:reminder -->
557
558<!-- SYNC:ai-mistake-prevention:reminder -->
559
560**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.
561
562<!-- /SYNC:ai-mistake-prevention:reminder -->
563
564<!-- SYNC:task-tracking-external-report:reminder -->
565
566- **MANDATORY** Bootstrap task tracking before target work; transition one task at a time.
567- **MANDATORY** Persist plan/review findings to `plans/reports/` incrementally and synthesize from disk.
568
569<!-- /SYNC:task-tracking-external-report:reminder -->
570
571<!-- SYNC:project-reference-docs-guide:reminder -->
572
573- **MANDATORY** Before investigating, planning, or coding, read `docs/project-config.json` (the project map: modules/paths, run-commands, conventions, architecture/workflow rules) + the required project-reference docs, and cite `Reference docs read: ...`.
574- **MANDATORY** Always include `lessons.md`; project config + conventions override generic framework defaults.
575- **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.
576
577<!-- /SYNC:project-reference-docs-guide:reminder -->
578
579<!-- SYNC:nested-task-creation:reminder -->
580
581- **MANDATORY** Parent workflow rows do not replace child phase tracking; expand phases and link the parent when nested.
582- **MANDATORY** Orchestrators pre-expand child skill phases before invocation; use `[N.M] $skill-name — phase` prefixes and one-`in_progress` discipline.
583
584<!-- /SYNC:nested-task-creation:reminder -->
585
586<!-- PROMPT-ENHANCE:STEP-TASK-CLOSING:START -->
587
588## Prompt-Enhance Closing Anchors
589
590**IMPORTANT MUST ATTENTION** follow declared step order for this skill; NEVER skip, reorder, or merge steps without explicit user approval
591**IMPORTANT MUST ATTENTION** for every step/sub-skill call: set `in_progress` before execution, set `completed` after execution
592**IMPORTANT MUST ATTENTION** every skipped step MUST include explicit reason; every completed step MUST include concise evidence
593**IMPORTANT MUST ATTENTION** if Task tools unavailable, maintain an equivalent step-by-step plan tracker with synchronized statuses
594
595<!-- PROMPT-ENHANCE:STEP-TASK-CLOSING:END -->
596
597<!-- SYNC:parallel-subagent-dispatch -->
598
599> **Parallel Sub-Agent Dispatch** — Plan parallelism the moment a task breakdown exists, BEFORE executing it — running provably independent tasks sequentially wastes wall-clock. Applies to every multi-step job: workflow steps, planning, batch updates, investigation, research, scans, reviews, doc sync. **Plan execution is metadata-gated, NEVER default-parallel** — fan-out follows ONLY what the plan declares (`PAR`/`SEQ` tags + per-phase write set); an untagged plan runs sequentially — why: a derived write set cannot see cascade or generated writes.
600>
601> 1. **Tag every task `PAR` or `SEQ`.** `PAR` = inputs exclude every pending task's output AND write set disjoint from every other `PAR`. Else `SEQ` — MUST ATTENTION name the dependency forcing it.
602> 2. **Group `PAR` into waves.** No edge between members. Two writers of one file NEVER share a wave. Read-only work (search, investigation, review, research) parallelizes freely.
603> 3. **Declare before dispatch:** `Parallel plan: wave 1 = [...] · wave 2 = [...] · SEQ = [...] (reason)`.
604> 4. **Spawn each wave in ONE message** — every `spawn_agent` call in one response, NEVER dripped per turn. Route each task to its specialist (`.claude/skills/shared/sub-agent-selection-guide.md`); NEVER `code-reviewer` as catch-all.
605> 5. **Brief each sub-agent self-contained:** goal · scope + owned files · reference docs · return contract (summary + `Full report:` path, per SYNC:subagent-return-contract) · incremental persistence to `plans/reports/` (per SYNC:incremental-persistence).
606> 6. **Barrier per wave.** Advance ONLY after EVERY member returns (a skipped conditional counts as returned). Merge, mark each task completed/skipped, THEN dispatch the next wave. Mutating steps wait for the barrier.
607> 7. **One level deep.** A dispatched sub-agent executes its own brief; further fan-out stays the orchestrator's job unless that agent's `.claude/agents/*.md` definition authorizes it.
608>
609> **NEVER parallelize:** tasks sharing a write target · a task consuming a pending task's output · trivial single-file work (dispatch overhead > gain) · an order a skill or workflow explicitly fixes · gates awaiting user approval.
610>
611> **Blocked until:** MUST ATTENTION every task tagged PAR/SEQ with a named reason per SEQ · waves declared + write-set disjointness checked · each wave spawned in ONE message · barrier honored before the next wave.
612
613<!-- /SYNC:parallel-subagent-dispatch -->
614
615<!-- SYNC:parallel-subagent-dispatch:reminder -->
616
617- **MANDATORY** After planning tasks, tag each PAR/SEQ and spawn every PAR wave as parallel sub-agents in ONE message — default parallel for workflows, batch updates, investigation, research, reviews; plan execution fans out ONLY on what the plan declares.
618- **MANDATORY** Disjoint write sets per wave · all-return barrier before the next wave · specialist routing · sub-agents NEVER fan out further unless their own agent definition authorizes it.
619
620<!-- /SYNC:parallel-subagent-dispatch:reminder -->
621
622<!-- SYNC:project-protocol-overlay -->
623
624> **Project Protocol Overlay** — Before executing this skill, resolve any PROJECT overlay rules layered onto it: match this skill's name against the `Target` column of the project's skill-protocol index (`docs/project-reference/skill-protocols-reference.md` by default; a `referenceDocs` entry in `docs/project-config.json` overrides the path), taking the most specific matching tier ONLY — exact name > glob > `*`. **That precedence orders overlays against EACH OTHER, never against this skill.** Read ONLY the matched bodies, resolved as `<protocols-dir>/<Name>.md`; a row's Body link is display text, never a read path. A matched body that is missing or malformed is REPORTED and skipped — never reconstructed from the index Description. No index, or no match -> proceed with no overlay, silently. Full contract: `.claude/skills/project-skill-protocol/references/registry.md`.
625>
626> Overlays are **ADDITIVE ONLY**: they ADD rules on top of this skill's own protocol and NEVER replace, override, disable, or reinterpret a rule it already states — removing every overlay must return this skill to exactly its documented behavior. An overlay is a BRIEF, not an authority escalation: it can NEVER waive a workflow gate, git discipline, a review gate, or a user-confirmation gate. A genuine overlay-vs-skill conflict, or two equally-specific overlays that directly contradict -> surface both to the user; NEVER resolve silently.
627
628<!-- /SYNC:project-protocol-overlay -->
629
630<!-- SYNC:project-protocol-overlay:reminder -->
631
632**MUST ATTENTION** resolve project protocol overlays for this skill BEFORE executing — most specific matching tier only (exact > glob > `*`, which ranks overlays against each other, NEVER against this skill), read only matched bodies at `<protocols-dir>/<Name>.md`; a missing or malformed body is reported, never reconstructed. Overlays are ADDITIVE ONLY (they never replace this skill's own rules) and are a brief, NEVER an authority escalation; an equal-specificity contradiction goes to the user.
633
634<!-- /SYNC:project-protocol-overlay:reminder -->
635
636## Closing Reminders
637
638**IMPORTANT MUST ATTENTION Goal:** Deliver a complete, prioritized map of every file relevant to the task — grep + graph combined — so downstream work starts with full coverage and zero blind spots.
639
640**IMPORTANT MUST ATTENTION — Protocols in force (concise digest of the SYNC/shared blocks this skill carries; each is a signpost to its canonical body above):**
641
642- **Graph-Assisted Investigation:** Run one graph command on key files before concluding.
643- **Incremental Persistence:** Append findings to a report file, never hold in memory.
644- **Subagent Return Contract:** Sub-agents return summary only, full report on disk.
645- **Nested Task Creation:** Expand child phases and link parent when nested.
646- **Project Reference Docs:** Read required project docs before target work; cite them.
647- **Task Tracking External Report:** Bootstrap task tracking, persist plan findings incrementally.
648- **Critical Thinking:** Traced proof per claim, confidence >80% to act.
649- **Evidence:** Cite `file:line`; speculation forbidden, <60% do not recommend.
650- **Cross-Service Check:** Scan producers, consumers, sagas, contracts for silent regressions.
651- **Rationalization Prevention:** Reject step-skipping evasions; show grep evidence.
652- **AI Mistake Prevention:** verify generated content against evidence, trace downstream references, verify all affected outputs, re-read after context loss, surface ambiguity.
653- **Parallel Sub-Agent Dispatch:** Tag tasks PAR/SEQ, group PAR into disjoint-write-set waves, spawn each wave in ONE message, barrier before advancing.
654
655**MUST ATTENTION** every protocol above is in force for this scout — honor its canonical body, not just the digest line.
656
657**IMPORTANT MUST ATTENTION** execute the FULL step pipeline in declared order — Phase 0 classify scope → Step 1 analyze request (entities/keywords/file-types) → Step 2 spawn parallel `scout` subagents → Step 3 graph-expand 2-3 key files (YOU) → Step 4 low-result check → Step 5 synthesize numbered prioritized list → ask the user directly next step; NEVER skip or merge a step — why: AI forgets the skill's own steps and ships a partial, under-searched map.
658
659**IMPORTANT MUST ATTENTION** run Phase 0 classification BEFORE spawning agents — scope (backend/frontend/full-stack) determines agent count; never spawn the default 3 when the prompt names one layer — why: extra agents waste budget and dilute focus
660**IMPORTANT MUST ATTENTION** graph expand is the MANDATORY step that finds what grep cannot — run at least ONE graph command (`connections`/`callers_of`/`trace --direction both`) on 2-3 key files when `.code-graph/graph.db` exists; NEVER skip it — why: structural relationships > text matches, and sub-agents cannot run graph — only you can
661**IMPORTANT MUST ATTENTION** cite `file:line` evidence for every claim. Confidence >80% to act, <60% = DO NOT recommend — why: speculation seeds blind spots downstream work inherits
662**MUST ATTENTION** stay in DISCOVERY lane — return prioritized file paths fast (3-5 min), no content deep-dives; that is `investigate`'s job — why: scope creep into analysis breaks the 3-5 min budget and duplicates the next step
663**MUST ATTENTION** if <5 files found, re-examine keywords and run a second pass with broader synonyms BEFORE synthesizing — why: a thin result is an under-searched result, not a small surface
664**MUST ATTENTION** run a post-grep graph trace whenever grep surfaces an entry-point file (entity, command, query, handler, controller, bus message, component, store, api-service) — trace reveals callers/consumers/event chains/tests grep cannot find — why: missing a downstream consumer = silent regression for the next step
665**MUST ATTENTION** spawn `scout`/`scout-external` subagents — NOT built-in `Explore` — why: only the custom agents carry graph CLI knowledge + Bash access
666**MUST ATTENTION** sub-agents return a SUMMARY only (≤10 finding bullets + `Full report:` path), writing full findings incrementally to `plans/reports/` — NEVER hold all results in memory or request full sub-agent output inline — why: context cutoff mid-run loses every in-memory finding; disk writes survive compaction
667**MUST ATTENTION** bootstrap task tracking BEFORE target work — the current task list first on context loss, one task `in_progress` at a time, expand child phases when nested under a workflow row — why: compaction wipes prior-work memory; resume from state, never duplicate
668**MUST ATTENTION** read required project-reference docs (always `lessons.md`; `domain-entities-reference.md` for business entities) BEFORE searching — project conventions override generic framework assumptions
669**MUST ATTENTION** use ask the user directly after completing (investigation workflow vs standalone, then $investigate vs $plan vs skip) — NEVER auto-proceed to the next step — why: the user owns scope; assuming standalone skips the workflow they wanted
670**MUST ATTENTION** for non-trivial bug/regression scouts, surface End-to-Start trace candidates (reader → storage → writer → consumer → producer) and enumerate every feeder path — why: starting at the first suspicious file collapses multiple producers into one false "flow"
671
672**Anti-Rationalization:**
673
674| Evasion | Rebuttal |
675| -------------------------------------- | --------------------------------------------------------------------------- |
676| "Graph step too slow, skip it" | Graph finds what 50 greps miss. NEVER skip when `graph.db` exists. |
677| "Only 2 files, no need for report" | Incremental write costs nothing. Skip = context loss risk. |
678| "Scope obvious, skip Phase 0" | Wrong agent set = missed files. Classify scope first, every time. |
679| "Already searched, results complete" | Show grep + graph evidence with `file:line`. No proof = incomplete. |
680| "Use Explore, it's a built-in" | Explore has NO graph awareness. Spawn `scout`/`scout-external` only. |
681| "<5 files, surface is just small" | Thin result = under-searched. Broaden synonyms and re-pass before synth. |
682| "I'll deep-dive these files now" | Discovery only — paths fast, no analysis. Deep-dive is `investigate`'s job. |
683| "Simple scout, skip workflow question" | User decides scope. NEVER assume standalone is acceptable. |
684
685**IMPORTANT MUST ATTENTION** Phase 0 classify first · graph expand is MANDATORY (never skip when `graph.db` exists) · cite `file:line` with confidence >80% — these three survive any long context, anchored top and bottom.
686
687**[TASK-PLANNING]** Before acting, analyze task scope and systematically break it into small todo tasks and sub-tasks using task tracking.
688
689<!-- CODEX:SYNC-PROMPT-PROTOCOLS:START -->
690
691## Hookless Prompt Protocol Mirror (Auto-Synced)
692
693Source: `.claude/.ck.json` + `.claude/skills/shared/sync-inline-versions.md` (`:full` blocks) + `.claude/scripts/lib/hookless-prompt-protocol.cjs`
694
695## [WORKFLOW-EXECUTION-PROTOCOL] [BLOCKING] Workflow Execution Protocol — MANDATORY IMPORTANT MUST CRITICAL. Do not skip for any reason.
696
697**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.
698
6991. **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.
7002. **ANALYZE:** Choose the best option: execute directly, invoke a skill, activate a standard workflow, or compose a custom step combination.
7013. **AUTO-SELECT:** Pick the best option yourself. Do not ask the user to choose between direct execution, skill, standard workflow, or custom workflow.
7024. **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.
7035. **CREATE TASKS:** task tracking for ALL workflow/skill/custom steps before execution when the selected path has multiple steps.
7046. **PARALLELIZE:** Before executing the task list, tag each task `PAR` (independent inputs + write set disjoint from every other `PAR` task) or `SEQ` (name the blocking dependency), group `PAR` tasks into waves, declare the wave plan, and spawn each wave's sub-agents in ONE message — all-return barrier per wave, fan-out one level deep unless a sub-agent's own definition authorizes further fan-out. Sequential-by-default is a defect when tasks are independent; do not parallelize shared write targets, output-consuming tasks, trivial single-file work, ordering a skill or workflow explicitly fixes, or user-approval gates.
7057. **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)
706
707## Shared AI-SDD Protocol Markers
708
709Source: `.claude/skills/shared/sync-inline-versions.md`
710
711## SYNC:ai-sdd-artifact-contract
712
713> **AI-SDD Artifact Contract** — Shared spec-driven development rules stay portable and source-owned.
714>
715> 1. Keep reusable AI-SDD principles in `.claude`; put repository-specific paths, commands, owners, products, and formats in project config/reference docs.
716> 2. Preserve cycle: `spec -> plan -> tasks -> implement -> verify -> update spec/docs`.
717> 3. Trace every requirement or invariant through decision, task, TC/test, source evidence, and docs/spec update.
718> 4. Treat code-to-spec extraction as reference-only until accepted by the canonical spec owner.
719> 5. Any supported AI tool may plan, implement, review, or verify with synced context; using multiple tools is optional.
720> 6. 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
721> 7. 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.
722>
723> **Active reference:** `shared/sdd-artifact-contract.md` in the active skills root.
724
725---
726
727## SYNC:ai-sdd-artifact-contract:reminder
728
729- **MANDATORY** Apply `shared/sdd-artifact-contract.md`; keep reusable AI-SDD in `.claude` and local rules in project docs.
730- **MANDATORY** Code-to-spec extraction is reference-only until canonical acceptance; any supported AI tool may execute with synced context.
731- **MANDATORY** Update `.claude` source before syncing generated mirrors; do not manually edit `.agents`, `.codex`, or `AGENTS.md`.
732- **MANDATORY** Missing or stale project config, root instruction files, or required reference docs route project-specific work through `$project-init` or the narrow setup route automatically.
733 **[TASK-PLANNING] [MANDATORY]** BEFORE executing any workflow or skill step, create/update task tracking for all planned steps, then keep it synchronized as each step starts/completes.
734
735## [LESSON-LEARNED-REMINDER] [BLOCKING] Task Planning & Continuous Improvement — MANDATORY. Do not skip.
736
737Break work into small tasks (task tracking) before starting. Add final task: "Analyze AI mistakes & lessons learned".
738
739**Extract lessons — ROOT CAUSE ONLY, not symptom fixes:**
740
7411. Name the FAILURE MODE (reasoning/assumption failure), not symptom — "assumed API existed without reading source" not "used wrong enum value".
7422. Generality test: does this failure mode apply to ≥3 contexts/codebases? If not, abstract one level up.
7433. Write as a universal rule — strip project-specific names/paths/classes. Useful on any codebase.
7444. Consolidate: multiple mistakes sharing one failure mode → ONE lesson.
7455. **Recurrence gate:** "Would this recur in future session WITHOUT this reminder?" — No → skip `$learn`.
7466. **Auto-fix gate:** "Could `$code-review`/`$code-simplifier`/`$security-review`/`$lint` catch this?" — Yes → improve review skill instead.
7477. BOTH gates pass → ask user to run `$learn`.
748 **[CRITICAL-THINKING-MINDSET]** Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence >80% to act.
749 **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.
750 **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.
751 **Goal-driven execution:** Define success criteria first, loop until verified, and stop only when observable checks pass.
752 **Tests verify intent:** Tests must protect business rules/invariants and fail when the protected intent breaks, not only mirror current behavior.
753
754## Common AI Mistake Prevention (System Lessons)
755
756- **Re-read files after context compaction.** Edit requires prior Read in same context; compaction wipes read state. Re-read before editing.
757- **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.
758- **Check downstream references before deleting.** Deletions cascade doc/code staleness. Map referencing files before removal.
759- **After memory loss, check existing state before creating new.** Compaction wipes prior-work memory. Query current state to resume — never blindly duplicate.
760- **Verify AI-generated content against actual code.** AI hallucinates APIs, class names, method signatures. Grep to confirm existence before documenting/referencing.
761- **Trace full dependency chain after edits.** Changing a definition misses downstream consumers. Trace the full chain.
762- **When renaming, grep ALL consumer file types.** Some file types silently ignore missing refs (no compile error). Search code, templates, configs, generated files.
763- **Trace ALL code paths when verifying correctness.** Code existing ≠ code executing. Trace early exits, error branches, conditional skips — not just happy path.
764- **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.
765- **Verify sub-agent results after context recovery.** Background agents may finish while parent compacted — grep-verify output, don't trust assumed completion.
766- **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.
767- **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).
768- **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.
769- **When debugging, ask "whose responsibility?" before fixing.** Trace caller (wrong data) vs callee (wrong handling). Fix at responsible layer — never patch symptom site.
770- **Test failure → record a provisional verdict before trace/edit, then investigate.** Use the full five-way taxonomy: SOURCE-WRONG (production violates intent), TEST-WRONG (assertion/setup is stale), TEST-NOT-OPTIMAL (valid but fragile or low-signal test), ENVIRONMENT-BLOCKED (external state prevents a verdict), or AMBIGUOUS (intent/evidence cannot choose safely). Then trace root cause and triangulate against the governing spec (`docs/specs/**` if one exists) AND source. NEVER weaken an assertion, add a skip, relax a timeout, or change source merely to force green.
771- **Grep ALL removed names after extraction/refactoring.** Primary file "done" ≠ secondary files clean. Grep entire scope for every removed symbol before declaring complete.
772- **Assume existing values are intentional — ask WHY before changing OR flagging one as a defect.** Pattern-matching as "wrong" skips context. Before changing or reporting any constant/limit/flag/cutoff: read comments, git blame, the CALLER's ordering (the guarantee that makes the value correct usually lives in code running immediately BEFORE the cited line), and 2+ sibling call sites of the same convention. A doc stating WHAT without WHY is missing rationale, not proof of a missing guard — and in a validation pass, an accurate `file:line` citation proves the transcription, never the defect.
773- **Verify ALL affected outputs, not just the first.** One build green ≠ all green. Multi-stack changes (backend/frontend/tests/docs) require verifying EVERY output.
774- **Evaluate fit before copying a nearby pattern.** Closest example ≠ matching preconditions — verify the new context shares the same constraints, base classes, scope, lifetime.
775- **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.
776- **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.
777- **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.
778- **[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.
779- **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".
780- **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.
781- **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.
782- **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.
783- **Assert the outcome your system OWNS, never the intermediate state your INFRASTRUCTURE owns.** When testing anything asynchronous (queue/broker delivery, retries, background jobs, caches, replication), assert the final business/entity state. NEVER assert the delivery bookkeeping — consume/send status, attempt counts, last-error, row existence or counts in a broker, scheduler, or outbox/inbox table. That bookkeeping lives in shared infrastructure that ANY co-running process (a peer worker, a second replica, a leftover local container) can write, usually under a deterministic shared key, so the assertion silently tests the developer's environment instead of the system: green when run alone, flaky the instant anything else shares that broker + database. Gate question for every assertion: "would this hold no matter WHICH process did the work?" — if no, assert the converged data state instead. Corollary: process-local fault injection and in-process telemetry cannot gate work any process may perform — use them as stress amplifiers (arm → bounded window → disarm → assert convergence), never as preconditions.
784- **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.
785
786<!-- CODEX:SYNC-PROMPT-PROTOCOLS:END -->