spec Skill | Agent Skills

Codex compatibility note:

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

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

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

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

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

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

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

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

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

Codex Project-Reference Loading (No Hooks)

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

Always read:

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

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

Situation-based docs:

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

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

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

Quick Summary

Portability: docs/specs/ is the fixed Feature Spec root. docs/templates/detailed-feature-spec-template.md remains the default template unless workflowPatterns.featureDocTemplate points to another template.

[IMPORTANT] task tracking — Break ALL work into small tasks BEFORE starting. For simple tasks, ask user whether to skip.

Goal: Own the entire Feature Spec lifecycle in one skill — author/maintain tech-free 8-section business Feature Specs (code evidence carried only in Section 8 test-case anchors, never prose), generate Section 8 test specifications, and reconcile those TCs with integration test code — producing a tech-free, AI-implementable Feature Spec whose Section 8 TC registry stays the single source of truth, traceable to integration test code, so any team can rebuild the feature on any stack from the spec alone. The mode you run determines which references/ body drives work; the shared §8 contract, M1-M6 mandates, and quality philosophy below apply every mode.

Summary:

Resolve the mode FIRST (draft/init/update/audit/amend/tests/sync) — an explicit [mode=<x>] wins, else infer from request + repo state; when ambiguous ask via a direct user question before any mutating mode, then read the matching references/{author,tests,sync}.md body and never run a mode from memory.
§1-7 prose is STRICTLY tech-free (no framework/product/language/persistence/auth names per spec-principles.md §3.2); technical identifiers live ONLY in evidence carriers — Section 8 is the canonical TC registry (TC-{FEATURE}-{NNN}) and must never be overwritten during update.
Every TC carries verifiable [Source: namespace/service/id] evidence — the sole exception is mode=draft, where idea-sourced specs use Evidence: TBD + provisional flag, upgraded to real anchors on the first code-sourced run.
Honor the M1-M6 mandates (sdd-artifact-contract.md) and canonical TC format (shared/tc-format.md); INDEX.md/ERD are derived artifacts — flag refresh need in update but never trigger $spec-index here.

Renamed: formerly /feature-spec (and earlier /feature-docs); the former /spec-tests skill is now folded in as mode=tests / mode=sync. Those names no longer resolve as slash commands — use $spec with the matching mode.

Modes (resolve mode FIRST — BLOCKING)

| Mode | Use when… | Body | | -------- | ----------------------------------------------------------------------------------------------------------------------------------------------- | ---------------------- | | draft | Author a provisional spec from an idea/requirement/prompt — no code yet (TDD-first, §8 Evidence: TBD, provisional marker) | references/author.md | | init | No docs/specs/{Bucket}/ exists — author a full 8-section spec from source | references/author.md | | update | Docs exist + code changed — section-impact-mapped updates | references/author.md | | audit | --audit flag or user asks — staleness report per section (never mutates docs) | references/author.md | | amend | [mode=amend] from the bugfix workflow — minimal regression-scoped §3/§4/§8 touch only | references/author.md | | tests | Generate or update Section 8 TC-{FEATURE}-{NNN} test specifications | references/tests.md | | sync | Reconcile §8 TCs ↔ integration test code (forward/reverse/harvest, orphan, staleness); harvest captures a SPEC-SILENT invariant into §4/§5/§8 | references/sync.md |

Mode resolution (do this before any work):

Parse the mode from the invocation: explicit [mode=<x>] arg wins; else infer from request + repo state ("from idea/requirements/prompt", "draft spec", "no code yet" → draft; no docs/specs/{Bucket}/ AND code exists to source from → init; docs exist + diff → update; "audit/stale" → audit; bugfix caller → amend; "write/update test specs", "TCs" → tests; "sync tests", "reconcile §8 with tests" → sync). draft vs init: both author a new spec, but draft sources from idea/requirement text (no code → Evidence: TBD, provisional) while init sources from existing code (real [Source:] evidence). "No docs" alone does NOT imply init — check whether code exists to source from. draft never auto-overwrites existing §8 TCs.
If ambiguous, present the detected mode via a direct user question before proceeding — NEVER auto-start a mutating mode.
Read the matching references/ body — it is the single source of truth for that mode's procedure, gates, and output contract. Do not run a mode from memory.

Key Rules (all modes):

[BLOCKING] Read docs/project-reference/spec-principles.md — repo-local prose/evidence rules (§3 prose scope + §3.2 banned prose-token list). For the AI-implementability criteria + tech-agnostic mandates, read .claude/skills/shared/sdd-artifact-contract.md ("AI-Implementability Gate" + mandates M1-M6) — those are the canonical authority, not the local stub.
[BLOCKING] EVERY test case MUST carry verifiable code evidence as a [Source: namespace/service/id] abstract anchor in its Section 8 hidden carrier — physical file:line lives only in the provenance sidecar.

Exception (mode=draft): an idea-sourced spec has no code yet — its §8 TCs carry Evidence: TBD (reference-only) and the spec is flagged provisional (provisional: true frontmatter + a "DRAFT — unverified until code lands" header banner). The first update/init run against real code MUST upgrade every TBD to a real [Source:] anchor and clear the provisional flag. This mirrors existing TDD-first handling — it relaxes evidence ONLY for draft, never for code-sourced modes.
[BLOCKING] Section 8 is the canonical TC registry — §8 TCs are the source of truth; integration test code implements them. The tests mode owns generation; sync mode reconciles drift; the author modes (draft/init) populate §8 at authoring time (draft with Evidence: TBD, init with real [Source:]) and MUST NOT overwrite existing TCs during UPDATE.
Authored docs MUST match the master template's 8 tech-free sections (Overview, Glossary, User Stories & AC, Business Rules, Domain Model, Process Flows, Permissions & Roles, Test Specifications) + YAML frontmatter — zero technical terms in prose, size caps enforced.
[BLOCKING] Canonical TC format authority: .claude/skills/shared/tc-format.md (GWT template, Evidence carrier, decade-numbering, Preservation Tests). M1-M6 mandates: .claude/skills/shared/sdd-artifact-contract.md — any violation FAILS the artifact.

docs/project-reference/feature-spec-reference.md — project-specific Feature Spec patterns (read directly when relevant). docs/project-reference/domain-entities-reference.md — domain entity catalog, relationships, cross-service sync.

8-Section Feature Spec Rules (canonical reference)

Canonical home for the Feature Spec rules; applies to any edit under the Feature Spec docs root.

Format: Tech-free 8-section Feature Spec. Activate the $spec skill before editing.

Read first: docs/project-reference/feature-spec-reference.md, docs/project-reference/spec-system-reference.md, and docs/project-reference/spec-principles.md. For behavior/public-contract changes, also read docs/project-reference/workflow-spec-test-code-cycle-reference.md.

8 sections (exact order): 1. Overview · 2. Glossary · 3. User Stories & Acceptance Criteria · 4. Business Rules · 5. Domain Model · 6. Process Flows · 7. Permissions & Roles · 8. Test Specifications — then a trailing Change History. No technical sections (Commands/Events/API/Cross-Service/Performance/Troubleshooting) — code is the technical source of truth.

Mandatory:

§1-7 prose is STRICTLY tech-free — no framework/product/language/persistence/messaging/auth names (banned tokens → spec-principles.md §3.2). Technical identifiers live ONLY in evidence carriers.
Section 5 (Domain Model): Mermaid ERD + [Source: component/{service}/{id}] abstract anchor per entity (cannot be omitted)
Section 4 (Business Rules): [Source: rule/{service}/{id}] abstract anchor per rule group
Section 8 (Test Specifications): canonical TC source — TC-{FEATURE}-{NNN} IDs, each carrying a hidden [Source: namespace/service/id] carrier + an IntegrationTest: field

Rules:

TC IDs live in Section 8 only — never authored in docs/specs/ directly
Section 8 authored via $spec [mode=tests]; $spec [mode=init] populates it only during initial authoring
Size caps: body (sections 1-7) ≤1200 lines, whole file ≤1800 (hard). Split the capability when body>1200 OR TCs>40
Change History entry required for every functional change (trailing section)

M1-M6 Compliance (BLOCKING — applies to every authored spec and every TC)

See .claude/skills/shared/sdd-artifact-contract.md → "AI-SDD Mandates (M1-M6)" for the full BLOCKING criteria. In brief: M1 tech-agnostic prose; M2 no source code in prose; M3 logical-IDs-first traceability with a SEPARATE [Source:] carrier; M4 AI-implementability (one interpretation, named success/failure); M5 rebuild-from-scratch on any stack from §1-8 prose alone. Tech terms are allowed ONLY inside evidence carriers (**Evidence**, IntegrationTest, [Source:]), YAML frontmatter, and ```mermaid ``` blocks.

Derived-Index Delegation

This skill owns the canonical Feature Spec (§1-8) and its §8 TC registry. The bucket INDEX.md and cross-capability ERD are derived artifacts regenerated by $spec-index FROM these specs — never a source of truth, and never authored here. In update mode, flag "derived spec artifact refresh may be required" but do NOT trigger $spec-index directly (separation of concerns).

Workflow Recommendation

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

Activate workflow-feature workflow (Recommended) — spec-driven with tests by default: scout → investigate → domain-analysis → why-review → spec → plan → plan-review → plan-validate → why-review → spec [mode=tests] → why-review → review-artifact --type=spec-tests → plan → plan-review → feature-implement → review-domain-entities → spec [mode=tests] → why-review → review-artifact --type=spec-tests → spec [mode=sync] → integration-test → integration-test-review → integration-test-verify → workflow-review-changes → production-readiness-review → security-review → changelog → test → docs-update → workflow-end → watzup

Execute $spec directly — run this skill standalone in the resolved mode

Next Steps

[BLOCKING] After completing, use a direct user question to present options. Do NOT skip — user decides:

"$spec [mode=tests] (Recommended)" — Generate/update Section 8 test specs for the documented features (if you just authored/updated a spec)
"$spec [mode=sync]" — Reconcile Section 8 TCs ↔ integration test code
"$review-artifact --type=spec-tests" — Audit TC coverage + GIVEN/WHEN/THEN quality
"Skip, continue manually" — user decides

Related Skills

| Skill | Relationship | When to Call | | ------------------------------------ | ---------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------ | | $spec-index | Derived consumer — assembles a regenerable navigation index/ERD FROM these Feature Specs (never a source of truth) | AFTER specs exist — (re)generate the bucket INDEX.md / cross-capability ERD over the canonical specs | | $review-artifact --type=spec-tests | Reviewer — audits TC coverage in Section 8 | After spec [mode=tests], to validate TC completeness and GIVEN/WHEN/THEN quality | | $integration-test | End consumer — generates test code from TCs in Section 8 | After spec [mode=tests], to produce actual integration test files | | $docs-update | Orchestrator — calls this skill as Phase 2 | Run $docs-update for full chain sync; it calls $spec internally | | $review-changes | Trigger — detects feature doc staleness | Calls $docs-update when a business doc is stale relative to code changes |

[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.

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

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: [...]."

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.

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

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

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

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.

Closing Reminders

IMPORTANT MUST ATTENTION Goal: Produce a tech-free, AI-implementable Feature Spec whose Section 8 TC registry stays the single source of truth, traceable to integration test code — so any team can rebuild the feature on any stack from the spec alone

Protocols in force (concise digest of the SYNC/shared blocks this skill carries — MUST ATTENTION honor each canonical body):

Cross-Service Check: ALWAYS scan producers, consumers, sagas, contracts before concluding; missing consumer = silent regression.
Evidence: cite file:line for every claim; confidence >80% to act, <60% NEVER recommend.
Critical Thinking: apply critical + sequential thinking; NEVER present a guess as fact.
AI Mistake Prevention: verify generated content against evidence, trace downstream references, verify all affected outputs, re-read after context loss, surface ambiguity.
IMPORTANT MUST ATTENTION [BLOCKING] Resolve the mode FIRST and read its references/{author,tests,sync}.md body — NEVER run draft/init/update/audit/amend/tests/sync from memory; ambiguous → a direct user question before any mutating mode — why: each mode's gates + output contract live in its body, not in this entry skill
IMPORTANT MUST ATTENTION [BLOCKING] EVERY test case MUST carry verifiable code evidence as a [Source: namespace/service/id] abstract anchor in its Section 8 hidden carrier — physical file:line → provenance sidecar only; sole exception mode=draft (Evidence: TBD + provisional flag, upgraded to real anchor on first code-sourced run) — why: a TC without evidence is unverifiable and silently rots
IMPORTANT MUST ATTENTION [BLOCKING] Section 8 is the canonical TC registry — existing TCs MUST NOT be overwritten during update; tests mode owns generation, sync mode reconciles drift — why: integration test code implements §8, so overwriting it orphans real tests
IMPORTANT MUST ATTENTION [BLOCKING] §1-7 prose is STRICTLY tech-free — no framework/product/language/persistence/messaging/auth names (banned tokens → spec-principles.md §3.2); technical identifiers live ONLY in evidence carriers, frontmatter, and mermaid blocks — why: M1/M5 require rebuild-from-scratch on any stack
IMPORTANT MUST ATTENTION [BLOCKING] Honor M1-M6 mandates (.claude/skills/shared/sdd-artifact-contract.md) + canonical TC format (.claude/skills/shared/tc-format.md) — any violation FAILS the artifact; size caps: body ≤1200, file ≤1800, split the capability when exceeded
IMPORTANT MUST ATTENTION INDEX.md/ERD are DERIVED — flag refresh need in update, NEVER trigger $spec-index here — why: separation of concerns keeps the canonical spec the only source of truth
IMPORTANT MUST ATTENTION evidence gate — cite file:line/grep for every claim, confidence >80% to act, <60% do NOT recommend; verify AI-generated TC/source anchors against ACTUAL code (grep to confirm) before authoring — why: hallucinated [Source:] anchors break traceability
IMPORTANT MUST ATTENTION cross-service check before concluding any spec/§8 work — scan producers, consumers, sagas, contracts; per touchpoint owner · message · risk (NONE/ADDITIVE/BREAKING) — why: a missing downstream consumer is a silent regression
IMPORTANT MUST ATTENTION [BLOCKING] Break work into small task tracking tasks BEFORE starting (one per file read) + a final review task; on context loss the current task list first, never duplicate — why: long spec files exhaust context and lose un-tracked progress
IMPORTANT MUST ATTENTION Search codebase for 3+ similar patterns and read existing spec siblings before authoring new content — match local conventions over generic defaults

Anti-Rationalization:

| Evasion | Rebuttal | | ------------------------------------------------ | -------------------------------------------------------------------------------------------- | | "Mode is obvious, skip the references/ body" | The body owns gates + output contract — running from memory drifts. Read it every time. | | "This TC's source is clear, skip the anchor" | No [Source:] carrier (or Evidence: TBD for non-draft) = unverifiable TC. Add the anchor. | | "update — just regenerate Section 8" | §8 is canonical; integration tests implement it. NEVER overwrite — sync reconciles drift. | | "One tech name in prose is harmless" | One banned token fails M1 and breaks rebuild-on-any-stack. Move it to an evidence carrier. | | "Small spec, skip task tracking" | Skip depth, NEVER skip tracking — context loss wipes un-tracked progress. | | "Index looks stale, I'll just run $spec-index" | Not this skill's job — flag the refresh need; derived artifacts regenerate separately. |

[TASK-PLANNING] MUST ATTENTION analyze task scope and break into small todo tasks/sub-tasks via task tracking before acting.

IMPORTANT MUST ATTENTION Resolve the mode FIRST + read its references/ body · EVERY §8 TC carries a [Source:] evidence anchor (except mode=draft) · §1-7 prose STRICTLY tech-free — the three rules this skill must never skip.

Hookless Prompt Protocol Mirror (Auto-Synced)

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

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

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

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

Shared AI-SDD Protocol Markers

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

SYNC:ai-sdd-artifact-contract

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

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

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

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

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

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

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

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

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

SYNC:ai-sdd-artifact-contract:reminder

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

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

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

Extract lessons — ROOT CAUSE ONLY, not symptom fixes:

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

Common AI Mistake Prevention (System Lessons)

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

Agent Skills: spec

Install this agent skill to your local

Skill Files