Agent Skills: workflow-code-to-spec

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.

Renamed: formerly workflow-build-specs — now $workflow-code-to-spec. The old name no longer resolves as a slash command.

Quick Summary

Goal: Keep the Feature Spec, implementation, tests, and project docs synchronized through a governed spec-driven workflow — a single entry point for spec-driven doc generation and maintenance over ONE canonical artifact, the tech-free 8-section Feature Spec at docs/specs/{Bucket}/README.{Feature}.md (code is the technical source of truth; there is no parallel engineering tree).

[SINGLE HOME] There is ONE canonical artifact — the tech-free 8-section Feature Spec authored by spec at docs/specs/{Bucket}/. There is no parallel A-E "Engineering Spec" bundle and no separate Business Feature Docs tree; spec-index only regenerates a DERIVED index/ERD over the Feature Specs. Authority: docs/project-reference/spec-system-reference.md.

One Canonical Artifact + Derived Aids

| Artifact | Path | Canonical? | Maintained By | | ---------------------- | ----------------------------------------- | -------------------------------- | ------------------------- | | Feature Spec | docs/specs/{Bucket}/README.{Feature}.md | Yes — single source of truth | spec | | Section 8 — Test Specs | Same file, Section 8 | Yes — canonical TC registry | spec [mode=tests] | | Bucket INDEX.md | docs/specs/{Bucket}/INDEX.md | Derived — regenerable | spec / spec-index | | System index / ERD | (generated on demand) | Derived — never canonical | spec-index (repurposed) |

App Bucket Mapping

Resolve service→bucket assignments from the canonical table in docs/project-reference/spec-system-reference.md → App Bucket Mapping — do not inline project-specific bucket names in this skill.

Mode Routing:

| Mode | When to Use | Step Sequence | | ----------- | -------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | init-full | Zero — no Feature Spec for target scope | scout → size-evaluation → plan → plan-review → plan-validate → spec [mode=init] → spec [mode=tests] → review-artifact --type=spec-tests → review-artifact → docs-update(final sync) → workflow-end → watzup | | update | Code changed, new requirement, new PBI | workflow-review-changes → spec [mode=update] → spec [mode=tests] → review-artifact --type=spec-tests → spec [mode=sync] → review-changes → docs-update(final sync) → workflow-end → watzup | | audit | Quarterly health check, verify doc freshness | scout → spec [mode=audit] → review-artifact → docs-update(final sync) → workflow-end → watzup |

Key Rules:

• MUST ATTENTION define success criteria before execution and loop until observable verification passes.
• MUST ATTENTION when creating/reviewing specs or tests, name Business Intent / Invariant Guarded or the protected business intent/invariant and ensure the test would fail if that intent breaks.
• Confirm mode via a direct user question BEFORE any action — NEVER skip Step 0
• Invoke skill invocation for EACH step — NEVER batch-complete or mark done without invocation
• Spawn sub-agents for 4+ capabilities in ONE message — NEVER sequential
• §1-7 of every Feature Spec are STRICTLY tech-free; the §5 Mermaid ERD is authored inside the Feature Spec — there is no separate ERD file
• Write findings incrementally after each section — NEVER hold in memory
• If shared skills/workflows/hooks/sync tooling changed, run npm run codex:sync before $workflow-end or record explicit N/A evidence; verify generated mirrors are current.

Step 0 — Mode Detection (MANDATORY FIRST)

Use a direct user question to confirm mode before any action.

Auto-detection rules:

IF docs/specs/{Bucket}/ has NO Feature Spec for the target scope
  → Suggest: init-full

IF git diff has service/frontend changes touching an already-spec'd capability
  → Suggest: update

IF explicit --audit flag OR user says "audit" / "check freshness" / "are docs stale"
  → Suggest: audit

Bucket + capability confirmation:

• Probe Glob docs/specs/{Bucket}/README.*.md to see which capabilities already have a Feature Spec.
• Map the changed services to a Bucket via the App Bucket Mapping table above.
• Confirm the capability name (PascalCase) with the user — this becomes README.{Feature}.md.

Present the detected mode with reasoning. User confirms before proceeding.

MODE: init-full

When to Use

Starting from zero: no docs/specs/{Bucket}/README.{Feature}.md for the target scope.

Step Sequence

## Step A — Discovery (scout)

$scout
  → Holistic codebase map — capability registry, entry points, integration boundaries
  → Identify the set of capabilities in scope (one Feature Spec each)
  → Task tracking: "scout — enumerate capabilities + entry points for {Bucket}"

## Step B — Size Evaluation and Divide-and-Conquer Planning (MANDATORY — runs BEFORE plan)

**[BLOCKING GATE]** Before writing any plan, evaluate scope and recursively decompose until
each capability is an independently authorable Feature Spec.

ESTIMATE:
  capability_count = count of distinct capabilities found in scout
  IF capability_count > 10:
    → SPLIT into groups: max 10 capabilities per run, run groups sequentially
  IF 4 ≤ capability_count ≤ 10:
    → Sub-agents mandatory (one spec sub-agent per capability)
  IF capability_count ≤ 3:
    → Single-session authoring

**Recursive decomposition rule:** If a single capability's Feature Spec would exceed the caps
(body §1-7 >1200 lines OR >40 TCs), split the capability into sibling specs.

Task tracking: "size-evaluation — count capabilities, decide split strategy"

## Step C — Plan

$plan
  → ONE task per capability Feature Spec (author §1-7 + §8 placeholder)
  → Core domain capabilities first, supporting ones last
  → Verify the current task list count ≥ capability_count before proceeding (BLOCKING gate)
  → Task tracking: "plan — produce per-capability authoring plan"

## Step D — Plan Review

$plan-review
  → Validate: every in-scope capability has an authoring task, no gaps
  → Verify: no sub-agent handles >3 capabilities; each capability within caps
  → Task tracking: "plan-review — validate capability coverage + caps"

## Step E — Plan Validation

$plan-validate
  → User confirms: bucket, capability list, scope, split strategy
  → Task tracking: "plan-validate — user confirms scope and split strategy"

$spec [mode=init]
  → Author the canonical tech-free 8-section Feature Spec PER capability:
      §1 Overview · §2 Glossary · §3 User Stories & AC · §4 Business Rules ·
      §5 Domain Model (Mermaid ERD — MANDATORY, authored INSIDE this file) ·
      §6 Process Flows · §7 Permissions & Roles · §8 Test Specifications
  → §1-7 STRICTLY tech-free; identifiers live only in §8 evidence carriers + `[Source: ns/service/id]` + ` ```mermaid ``` ` blocks
  → Output: docs/specs/{Bucket}/README.{Feature}.md + bucket INDEX.md
  → Sub-agents for 4+ capabilities (BLOCKING: ONE message spawn); each prompt includes capability name, output path, tech-agnostic contract, SYNC protocols
  → Caps: body §1-7 ≤1200 lines, whole file ≤1800 (hard) — split when body>1200 OR TCs>40

$spec [mode=tests]
  → Populate Section 8 with TC-{FEATURE}-{NNN} entries (BDD, Business Intent, abstract Evidence, IntegrationTest field, Status)

$review-artifact --type=spec-tests
  → Review §8 TCs for invariant coverage, GIVEN/WHEN/THEN completeness, duplicate TC IDs

$review-artifact
  → Quality check the Feature Spec(s):
    - §1-7 strictly tech-free (zero framework/product/language terms in prose)
    - §5 Mermaid ERD present with entities, relationships, cardinalities
    - §8 every TC has Business Intent, abstract `[Source: ns/service/id]` anchor, IntegrationTest field, Status
    - YAML frontmatter present; body ≤1200 / whole file ≤1800 lines
  → PASS criteria: zero [UNVERIFIED] without exclusion reason + zero tech terms in §1-7
  → Gap found → validate findings → fix validated gaps → restart full review-artifact pass from the first check

$docs-update
  → Near-final synchronization sweep across project docs, the Feature Spec(s), and Section 8
  → MUST run after review-artifact fixes and before $workflow-end
  → (Optional) regenerate the derived bucket INDEX / ERD via $spec-index
  → Report skipped sub-phases explicitly when no impacted docs exist

$workflow-end

$watzup
  → Session summary: capabilities authored, files written, ~lines, §8 TC counts, coverage gaps, open questions (confidence < 80%), plus final $understand handoff

MODE: update

When to Use

Code changed (new feature, bug fix, refactor, new PBI). Sync the Feature Spec incrementally.

Scope Sources

1. Auto-detect from git diff --name-only HEAD (default)
2. Explicit capability list from user
3. New PBI or requirement description (map to affected capabilities manually)

Step Sequence

$workflow-review-changes
  → Full code review cycle + docs-update (Phase 2 spec diff-scoped sync)
  → Produces: impact map (capabilities affected, Feature Spec sections to update)

$spec [mode=update]
  → Update only the impacted sections of each affected Feature Spec (full 8-section pass with 3-pass verification when restructuring)
  → SKIP if docs-update Phase 2 completed with zero gaps — mark "Skipped: docs-update Phase 2 sufficient"

$spec [mode=tests]
  → **EXPLICIT TC STEP — required when new functionality added**
  → SKIP if changes are purely cosmetic (CSS, comments, config) — mark "Skipped: no behavioral impact"
  → Mode detection: new commands/queries/endpoints → implement-first; PBI-driven → TDD-first; TC edits only → update
  → Write/update TCs in Section 8; grep existing IDs before assigning; never overwrite Tested TCs

$review-artifact --type=spec-tests
  → Review updated/planned TCs for invariant coverage, GIVEN/WHEN/THEN completeness, stale TC handling, duplicate IDs
  → SKIP if $spec [mode=tests] skipped — mark "Skipped: no TC changes"
  → BLOCK sync if review finds missing invariants, ambiguous behavior, or TC/code/spec drift

$spec [mode=sync]
  → Refresh the derived bucket INDEX TC counts from Section 8 (Section 8 stays canonical — no separate dashboard)
  → SKIP if no TC changes in this cycle

$review-changes
  → Holistic review of Feature Spec changes
  → Verify: spec changes match code changes (no over/under documentation); §8 covers all new functionality

$docs-update
  → Near-final synchronization sweep across project docs, the Feature Spec(s), and Section 8
  → MUST run after review-changes fixes and before $workflow-end
  → Report skipped sub-phases explicitly when no impacted docs exist

$workflow-end

$watzup
  → Summary: capabilities updated, Feature Spec sections changed, TCs added/updated, new open questions, plus final $understand handoff

New PBI / Requirement Update Protocol

Triggering update from new PBI or requirement (not code change):

After $workflow-review-changes:
  → User provides PBI text or requirement description
  → Map requirement → affected domain entities → affected capabilities
  → $dor-gate: required when a new/changed PBI is being made implementation-ready; PASS or WARN before planned specs/TCs become guidance
  → $pbi-mockup: required when the PBI changes user-facing UI/journeys; skip with reason for backend-only requirements
  → Treat as "speculative update": add new rules/sections to the Feature Spec marked [PLANNED — not yet implemented]
  → Add corresponding TCs in Section 8 marked Status: Planned
  → $review-artifact --type=spec-tests: review planned TCs before refreshing the derived index
  → These become implementation guidance, not verified spec

MODE: audit

When to Use

Periodic health check (quarterly or before major release). Verify the Feature Specs are current.

Audit Time Estimation

audit_effort = capability_count × 8min_per_capability
example: 8 capabilities × 8min = ~1h (AI-assisted)

Budget multiplier: If last audit was >90 days ago → ×1.5 (more drift expected).

Step Sequence

$scout
  → Quick codebase scan: current state of entities, commands, controllers (lightweight, 30min max)

$spec [mode=audit]
  → Compare each Feature Spec's last_updated vs git log of the source it documents
  → Output: stale capabilities/sections table with recommended update scope

$review-artifact
  → Consolidated audit report
  → Produce: plans/reports/spec-audit-{date}-{Bucket}.md
  → Include: total stale coverage %, estimated update effort, priority order
  → (Optional) $spec-index [mode=audit] — report which DERIVED index/ERD aids lag their source specs

$docs-update
  → Near-final synchronization sweep; MUST run after review-artifact conclusions and before $workflow-end
  → Report skipped sub-phases explicitly when no impacted docs exist

$workflow-end

$watzup
  → Present action plan: which capabilities to update first
  → Recommend: run update mode scoped to stale capabilities
  → Run final $understand handoff

Conditional Skip Rules

| Step | Skip When | | ---------------------------------------------- | -------------------------------------------------------------------------------------------------------------- | | §5 Mermaid ERD in init | Never — the ERD is a mandatory section of the Feature Spec | | $spec [mode=tests] in init | User explicitly requests a behavior-doc-only pass (TCs deferred to a later cycle) | | $dor-gate in update | Update source is code diff only, existing PBI is already DoR-ready, or no PBI readiness decision is being made | | $pbi-mockup in update | Backend-only/non-UI requirement, code diff only, or existing mockup already covers the change | | $review-artifact --type=spec-tests in update | $spec [mode=tests] skipped because there were no TC changes | | $spec [mode=sync] | No TC changes in this update cycle | | $docs-update near-final sync | Never skip entirely; sub-phases may be skipped only with explicit reason in the docs-update report | | $review-artifact audit pass | No stale specs found AND no UNVERIFIED items | | $spec-index (derived) | No derived index/ERD is maintained for this bucket, or it is already current |

Sub-Agent Coordination Protocol (init-full, 4+ capabilities)

1. $scout + $plan in main context → capability registry + per-capability task list
2. Spawn spec sub-agents (one per capability) in ONE message
3. Wait for all sub-agents to complete
4. Spawn spec [mode=tests] sub-agents (one per capability) in ONE message to populate Section 8
5. Main context assembles + verifies in $review-artifact

Each spec sub-agent receives:

• Capability name + bucket
• Output path: docs/specs/{Bucket}/README.{Feature}.md
• Tech-agnostic contract (§1-7 tech-free; §5 ERD mandatory)
• Incremental persistence instruction (write after each section)

Integration with docs-update workflow step

docs-update called as a workflow step (not standalone) synchronizes the single canonical artifact:

docs-update (as workflow step):
  Phase 1: Project docs (inline — unchanged)
  Phase 2: $spec update mode (impacted sections of the Feature Spec)
  Phase 3: $spec [mode=tests] (Section 8 TCs)
  Phase 3.5: $review-artifact --type=spec-tests (required when Phase 3 changes TCs)
  Phase 4: $spec [mode=sync] (refresh derived INDEX TC counts)
  Phase 4.5: $spec-index (OPTIONAL — regenerate derived bucket INDEX / ERD if one is maintained)

The Feature Spec stays in sync on every feature/bugfix/refactor workflow.

When to Use vs When NOT to Use

Use This Workflow

• First-time Feature Spec authoring for a capability or full bucket
• After significant code changes (new feature, major refactor)
• Onboarding new team to a capability — the Feature Spec as knowledge handoff artifact
• Before tech migration or re-implementation (pair with a derived reimplementation guide)
• Quarterly spec health audits
• After new PBIs groomed for implementation
• Compliance documentation — prove system behavior in plain language
• Verify design intent — compare the Feature Spec against original vision

Use Standalone Skills Instead

| Goal | Use | | --------------------------------------------------- | ----------------------------- | | Update one specific Feature Spec after small change | $spec directly | | Add/sync Section 8 TCs | $spec [mode=tests] directly | | Regenerate a derived bucket index / ERD | $spec-index directly | | Understand one specific feature | $investigate | | Write integration tests from existing Section 8 | $integration-test |

[TASK-PLANNING] Before acting, analyze task scope and systematically break it into small todo tasks and sub-tasks using task tracking.

IMPORTANT MANDATORY Steps: $scout -> $plan -> $plan-review -> $plan-validate -> $spec -> $spec [mode=tests] -> $review-artifact --type=spec-tests -> $review-artifact -> $docs-update -> $workflow-end -> $watzup

[BLOCKING] Invoke skill invocation for EACH step — NEVER batch-complete, NEVER mark done without skill invocation. [BLOCKING] Confirm mode via a direct user question BEFORE any action — NEVER skip Step 0. [BLOCKING] Spawn sub-agents for 4+ capabilities in ONE message — NEVER sequential spawning. [BLOCKING — Context Compaction / Session Resume] At any session start or after context compaction: (1) the current task list FIRST — resume existing, NEVER create duplicates; (2) re-glob docs/specs/{Bucket}/ to see which capabilities already have a Feature Spec — skip those; (3) NEVER re-run $scout or $plan in a resumed session. [BLOCKING] Read docs/project-reference/spec-principles.md before running any author/update/audit step — it is the shared spec quality baseline (tech-agnostic rule + banned-token list).

Goal Contract propagation (workflow-owned): At workflow start, resolve the active Goal Contract per SYNC:goal-contract-satisfaction-loop (active plan goal.md → plans/goals/{YYMMDD-HHmm}-{slug}/goal.md → create from the spec request). Map each spec/test/code cycle output (Feature Specs authored, TCs written, audit findings fixed) to the saved success criteria and append the evidence to the goal file's Iteration Log per cycle. Before $workflow-end, emit the Goal Satisfaction matrix (PASS/FAIL/BLOCKED); completion requires every required criterion PASS or BLOCKED with a user-facing escalation.

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

1. Call the current task list first. If a matching active parent workflow row exists, set nested=true and record parentTaskId; otherwise run standalone.
2. Create one task per declared phase before phase work. When nested, prefix subjects [N.M] $skill-name — phase.
3. When nested, link the parent with TaskUpdate(parentTaskId, addBlockedBy: [childIds]).
4. Orchestrators must pre-expand a child skill's phase list and link the workflow row before invoking that child skill or sub-agent.
5. Mark exactly one child in_progress before work and completed immediately after evidence is written.
6. 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.

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.

Incremental Result Persistence — MANDATORY for all sub-agents or heavy inline steps processing >3 files.

1. Before starting: Create report file plans/reports/{skill}-{date}-{slug}.md
2. After each file/section reviewed: Append findings to report immediately — never hold in memory
3. Return to main agent: Summary only (per SYNC:subagent-return-contract) with Full report: path
4. 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.

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

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

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

Closing Reminders

IMPORTANT MUST ATTENTION Goal: Ensure the Feature Spec, implementation, tests, and project docs stay synchronized through a governed spec-driven workflow.

Protocols in force (concise digest of the SYNC/shared blocks this skill carries):

• Nested Task Creation: expand child phases, link parent when nested; NEVER batch transitions.

• Critical Thinking: traced proof per claim, confidence >80% to act; NEVER present 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.

• Incremental Persistence: persist findings to plans/reports/ per section; NEVER hold in memory.

• Sub-Agent Return Contract: sub-agents return summary only; NEVER request full output inline.

• [BLOCKING] Confirm mode via a direct user question BEFORE any action — NEVER skip Step 0

• [BLOCKING] Invoke skill invocation for EACH step — NEVER batch-complete or mark done without invocation

• [BLOCKING] Spawn sub-agents for 4+ capabilities in ONE message — NEVER sequential spawning

• [BLOCKING] ONE canonical artifact — the Feature Spec at docs/specs/{Bucket}/README.{Feature}.md; the §5 Mermaid ERD is authored INSIDE it (no separate ERD file, no A-E tree)

• [BLOCKING] Scout holistically FIRST — capability registry MUST exist before plan creation; NEVER re-run scout or plan in a resumed session

• [BLOCKING] Plan decomposes big→small — ONE task per capability Feature Spec; split a capability when body §1-7 >1200 lines OR TCs >40

• [BLOCKING] Per-section authoring: write each section immediately — NEVER accumulate across sections

• [REQUIRED] §1-7 STRICTLY tech-free (no framework names, no language constructs, no class names in prose); identifiers live only in §8 evidence carriers, [Source: ns/service/id], and ```mermaid ``` blocks — mark [UNVERIFIED] not blank

• [REQUIRED] Each sub-agent prompt MUST include: capability name, output path, tech-agnostic contract, SYNC protocols (critical-thinking, evidence-based, incremental-persistence, cross-scope boundary)

• [BLOCKING] Context compaction / session resume → the current task list first, re-glob existing Feature Specs, skip done capabilities — NEVER re-run scout or plan

• [BLOCKING] review-artifact: PASS = zero [UNVERIFIED] without exclusion reason + zero tech terms in §1-7; gap found → validate findings → fix validated gaps → restart the full review-artifact pass from the first check

• [BLOCKING] Verify the current task list count ≥ capability_count before any authoring begins — this is the plan completeness gate

• [REQUIRED] Apply critical thinking — every claim needs traced proof, confidence >80% to act. Anti-hallucination: never present guess as fact.

• [REQUIRED] Apply AI mistake prevention — holistic-first debugging, fix at responsible layer, surface ambiguity before coding, re-read files after compaction.

[IMPORTANT] Analyze how big the task is and break it into many small todo tasks systematically before starting — this is very important. Anti-Rationalization:

| Evasion | Rebuttal | | --------------------------------------- | ----------------------------------------------------------------------------------------------------------- | | "Purpose obvious" | Anchor it anyway — primacy/recency keeps outcome active through long prompts. | | "Existing reminders enough" | Echo Goal in Closing Reminders — bottom anchor prevents drift. | | "Skip evidence for prompt edits" | Cite changed file evidence and verify no stale protocol text remains. | | "Re-extract the A-E engineering bundle" | Not part of the spec model. ONE Feature Spec; the ERD is §5. spec-index only regenerates a derived index. |

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.

1. 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.
2. ANALYZE: Choose the best option: execute directly, invoke a skill, activate a standard workflow, or compose a custom step combination.
3. AUTO-SELECT: Pick the best option yourself. Do not ask the user to choose between direct execution, skill, standard workflow, or custom workflow.
4. 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.
5. CREATE TASKS: task tracking for ALL workflow/skill/custom steps before execution when the selected path has multiple steps.
6. 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.

1. Keep reusable AI-SDD principles in .claude; put repository-specific paths, commands, owners, products, and formats in project config/reference docs.
2. Preserve cycle: spec -> plan -> tasks -> implement -> verify -> update spec/docs.
3. Trace every requirement or invariant through decision, task, TC/test, source evidence, and docs/spec update.
4. Treat code-to-spec extraction as reference-only until accepted by the canonical spec owner.
5. Any supported AI tool may plan, implement, review, or verify with synced context; using multiple tools is optional.
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
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.

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:

1. Name the FAILURE MODE (reasoning/assumption failure), not symptom — "assumed API existed without reading source" not "used wrong enum value".
2. Generality test: does this failure mode apply to ≥3 contexts/codebases? If not, abstract one level up.
3. Write as a universal rule — strip project-specific names/paths/classes. Useful on any codebase.
4. Consolidate: multiple mistakes sharing one failure mode → ONE lesson.
5. Recurrence gate: "Would this recur in future session WITHOUT this reminder?" — No → skip $learn.
6. Auto-fix gate: "Could $code-review/$code-simplifier/$security-review/$lint catch this?" — Yes → improve review skill instead.
7. 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.