Sync Skills Shared Protocols

---
name: sync-skills-shared-protocols
description: '[Skill Management] Use when shared protocol checklists change and need propagating across 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:START -->
## Codex Project-Reference Loading (Hook-Independent)

Claude and Codex use static project-reference loading as the authority; hooks may accelerate discovery but never replace the explicit read.
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.
<!-- CODEX:PROJECT-REFERENCE-LOADING:END -->

## Quick Summary

**Goal:** Two operations — (A) propagate updated content for existing SYNC: blocks across all skills, or (B) add a new SYNC: block to all skill/agent files that don't have it yet.

**Summary:** Use `$sync-skills-shared-protocols` for Operation A (update existing blocks) or Operation B (add a new tiered block); `/sync-protocols` no longer resolves.

**Canonical source:** `.claude/skills/shared/sync-inline-versions.md`

**Key Rules:** Edit the canonical source first; change only SYNC block bodies; sync a `:reminder` block the same mechanical way — by passing the `{tag}:reminder` tag to the script — and only when asked; use the script for bulk insertion; verify balance, parity, and the computed target inventory.

**Workflow:** Choose Operation A or B → follow its canonical-source and target-inventory steps → verify before reporting.

## Workflow

### Operation A: Update Existing Block Content

Use when a SYNC: block already exists in files and push updated content to all of them.

### Step 1: Identify What Changed

If user specifies a tag name (e.g., `sync-skills-shared-protocols understand-code-first`):

- Sync only that tag

If no tag specified:

- Read `sync-inline-versions.md`
- Ask user which tag(s) to sync, or "all"

### Step 2: Read Canonical Content

For each tag to sync:

1. Read `.claude/skills/shared/sync-inline-versions.md`
2. Extract content under `## SYNC:{tag-name}` heading (everything between that heading and the next `---` or `## SYNC:`)

### Step 3: Find All Skills with Tag

```bash
grep -rl "SYNC:{tag-name}" .claude/skills/*/SKILL.md
```

### Step 4: Replace Content in Each Skill

For each file found:

1. Find `<!-- SYNC:{tag-name} -->` open tag
2. Find `<!-- /SYNC:{tag-name} -->` close tag
3. Replace everything between them with the canonical content
4. Do NOT touch `:reminder` blocks while syncing the plain tag — they are a SEPARATE fence pair with their own canonical section, and the script never crosses between them (`sync-update-blocks.py:64-66`). Sync one by passing `{tag}:reminder` as its own tag.

### Step 5: Verify

Run these checks after all replacements:

```python
# 1. SYNC tag balance (all opens have matching closes)
# 2. Content matches canonical source
# 3. No content outside SYNC blocks was modified
```

Report:

- Tags synced
- Files updated (count)
- Any balance issues

### Step 6: Reminder Blocks

`:reminder` blocks (`<!-- SYNC:{tag}:reminder -->`) are 1-line summaries at the bottom of skills for AI recency attention (Primacy-Recency).

**Where a canonical `:reminder` section exists, they ARE mechanically syncable — do NOT hand-edit those.** `sync-update-blocks.py` treats `{tag}:reminder` as an ordinary tag: it reads the `## SYNC:{tag}:reminder` section from the canonical source and replaces the matching fence pair (`sync-update-blocks.py:11-14,25-35`). Hand-writing such a reminder desynchronizes it from the canonical text the very next time the script runs.

> **Coverage is PARTIAL — check before you rely on the script.** The canonical source does NOT yet carry a `:reminder` section for every reminder tag in use: roughly half the live reminder tags have no canonical section, and running the script on one of those fails hard with `ERROR: section not found` (`sync-update-blocks.py:33`) rather than doing nothing. Confirm the section exists first:
>
> ```bash
> grep -n "^## SYNC:{tag}:reminder" .claude/skills/shared/<canonical-source>.md
> ```
>
> **Section present** → sync it with the script; never hand-edit. **Section absent** → the script cannot serve you at all. Add the canonical `## SYNC:{tag}:reminder` section first and then sync, so the reminder becomes managed like the rest. Do not silently hand-edit the fenced body as a workaround: that leaves a fenced block the script believes it owns and will overwrite the moment the section appears.

What is true is that a reminder is a SEPARATE tag from its parent, not a shorter view of it: syncing `foo` never touches `foo:reminder`, because the script's fence regexes require whitespace before the closing marker and so cannot cross between the two (`sync-update-blocks.py:64-66`). Update a reminder by running the script on `{tag}:reminder` explicitly, after editing that section in the canonical source — and only when the user asks for reminders too.

```bash
py -3 .claude/scripts/sync-update-blocks.py --dry-run {tag}:reminder   # Windows
python3 .claude/scripts/sync-update-blocks.py {tag}:reminder           # macOS/Linux
```

### Step 7: OVERRIDE Blocks — the sanctioned divergence

Not every carrier can take the canonical body verbatim. A review skill that must route its
sub-agents to a DIFFERENT specialist needs the same protocol substance with a different
`agent_type`. That is what `<!-- OVERRIDE:{tag} -->` … `<!-- /OVERRIDE:{tag} -->` is for.

**The contract:**

- **`sync-update-blocks.py` does NOT touch an OVERRIDE block.** It is an intentional per-skill
  divergence, so the equality property that binds `SYNC:` carriers is deliberately not applied.
- **Divergence is limited to ROUTING, not substance.** The `sync-carrier-parity` suite's
  OVERRIDE-SUBSTANCE GUARD pins every OVERRIDE copy to the canonical protocol COUNT and to each
  protocol's header AND body verbatim; only the Subagent-Type / Agent-Call / Reference-Docs
  sections may differ. Silent staleness on substance is the failure mode it exists to catch.
- **The carrier set is pinned.** The guard asserts the exact number of OVERRIDE carriers, so a
  new one appearing — or an existing one vanishing — fails the suite rather than passing quietly.
- **Both markers are recognized as fences.** `check-subagent-routing.cjs` treats `SYNC` and
  `OVERRIDE` openers/closers identically for balance checking.

**Live carriers (3 skills × 2 tags):** `architecture-review`, `integration-test-review`, and
`ui-review` each override `fresh-context-review` and `review-protocol-injection`.

**Maintaining one:** edit the canonical section, run the script for the `SYNC:` carriers, then
**hand-merge** the same substance change into each OVERRIDE block, preserving its
`agent_type` customization. The guard tells you if you missed one.

**Do NOT reach for OVERRIDE to avoid a sync conflict.** It is for a carrier that genuinely must
dispatch elsewhere. Any other divergence belongs in the canonical source, so every carrier gets it.

---

### Operation B: Add a New Tiered Block

Use when a NEW SYNC: block needs tiered propagation. Derive the on-disk target inventory at runtime; never copy a fixed skill or agent count into this contract.

- `SKILL_BLOCK_ORDER` is the base tier for every skill.
- `ORCHESTRATOR_SKILL_BLOCK_ORDER` extends that base only for skills in `ORCHESTRATOR_SKILLS`.
- Agent tiers remain independently governed by the injector's explicit agent sets.

#### Agent quality parity and skill connections

When the new block applies to work performed by a sub-agent, update the canonical agent matrix before injection:

1. Add the applicable quality tag(s) to `AGENT_QUALITY_BLOCKS` in `.claude/scripts/agent_protocol_matrix.py`; keep orchestration-only blocks out of leaf agents.
2. Confirm every agent has a valid `AGENT_SKILL_CONNECTIONS` entry to its canonical task skill(s), and add the carrier owner for any newly propagated review/test protocol.
3. Run `py -3 .claude/scripts/agent_protocol_matrix.py --validate`, then run both `inject_agent_protocol_blocks.py` and `inject_agent_skill_connections.py` (dry-run first, real run second).
4. Verify the generated `AGENT-SKILL-CONNECTIONS` sections and the agent-universal-rules suite before regenerating mirrors.

The connection map is explicit routing metadata. It links the agent prompt to the canonical skill procedure while the matrix injects only applicable quality blocks; do not blanket-copy skill bodies into a leaf agent when that would import caller-owned orchestration.

This is a bulk-insert operation, not a content update. Verify the computed target set before writing so orchestration-only rules never leak into every skill.

**When to use:** A new protocol rule is added to `.claude/skills/shared/sync-inline-versions.md` and should appear in static carriers (`CLAUDE.md`, `AGENTS.md`, Codex, skills, and agents).

#### Step B1: Add block content to canonical source

Edit `.claude/skills/shared/sync-inline-versions.md` and add a new section:

```markdown
## SYNC:{new-block-name}

> **[Rule content here]**

---
```

#### Step B2: Add block to `sync-hooks-to-skills.py`

Edit `.claude/scripts/sync-hooks-to-skills.py`:

1. Add entry to `BLOCKS` dict:

```python
BLOCKS = {
    # ... existing blocks ...
    "new-block-name": """\
<!-- SYNC:new-block-name -->

> **[Full block content here — exactly as it should appear in files]**

<!-- /SYNC:new-block-name -->""",
}
```

2. Add 1-line reminder to `REMINDERS` dict:

```python
REMINDERS = {
# ... existing reminders ...
"new-block-name": """\

> **[IMPORTANT]** Use task tracking to break ALL work into small tasks BEFORE starting.

<!-- SYNC:new-block-name:reminder -->
**MUST ATTENTION** [one-line summary of the rule].
<!-- /SYNC:new-block-name:reminder -->""",
}
```

3. Add the block name to the relevant tier list(s) (controls which targets receive it + insertion order):

```python
# Skills-3: the two universal blocks + the orchestration block. Do NOT add
# agent-only rules here. `parallel-subagent-dispatch` is deliberately SKILL-ONLY
# — not because an agent cannot spawn, but because an agent receives ONE brief an
# orchestrator has ALREADY partitioned, so re-running the partitioning protocol in
# the leaf re-decides an upstream decision. It is ALSO declared in
# agent_protocol_matrix.py EXCLUDED_ORCHESTRATION and in the TC-UAR-017
# AGENT_ADOPTION_EXEMPT set (see "Skill-only blocks" below).
SKILL_BLOCK_ORDER = ["critical-thinking-mindset", "ai-mistake-prevention",
                     "parallel-subagent-dispatch"]

# Core-6: every agent (skills/SKILL.md is unaffected).
CORE_BLOCK_ORDER = ["critical-thinking-mindset", "ai-mistake-prevention",
                    "sequential-thinking-protocol", "task-tracking-external-report",
                    "project-reference-docs-guide", "agent-bootstrap"]

# Code-10: Core-6 + code-investigation blocks, for agents that read/review AND fix code.
CODE_BLOCK_ORDER = CORE_BLOCK_ORDER + ["understand-code-first", "evidence-based-reasoning",
                                       "cross-service-check", "fix-layer-accountability"]

# Readonly-Code-8: Core-6 + reading-discipline blocks only, for read-only/design
# agents that locate/read/design code but never fix a layer or cross a service
# boundary (excludes the two mutation-oriented blocks).
READONLY_CODE_BLOCK_ORDER = CORE_BLOCK_ORDER + ["understand-code-first", "evidence-based-reasoning"]
```

**Agent tiering:** agents no longer share one block list. `find_target_files()` classifies each `.claude/agents/*.md` by explicit membership in one of three sets:

- `CODE_AGENTS` (17 code/review/fix agents → `CODE_BLOCK_ORDER`, Code-10).
- `READONLY_CODE_AGENTS` (2 read-only/design agents — `researcher`, `ui-ux-designer` → `READONLY_CODE_BLOCK_ORDER`, Core-6 + understand-code-first + evidence-based-reasoning; the mutation-oriented `cross-service-check` + `fix-layer-accountability` are deliberately excluded to save tokens on agents that only locate/read/design code).
- `CORE_ONLY_AGENTS` (8 non-code agents → `CORE_BLOCK_ORDER`, Core-6).

An agent in **none of the three sets (or in more than one)** raises `SystemExit` — no silent default; classify it before the script will run. Skills always use `SKILL_BLOCK_ORDER`. Pass `--agents-only` to scope a run to agents (skip skills).

> **Adding a new agent:** add its basename to exactly one of `CODE_AGENTS` / `READONLY_CODE_AGENTS` / `CORE_ONLY_AGENTS` in `sync-hooks-to-skills.py` **and** in the regression suite `.claude/hooks/tests/suites/agent-universal-rules.test.cjs` (TC-UAR-005 fails until both agree). The two enforce one invariant.
>
> **Note — the inserter is insert-only.** Moving an agent from CODE to READONLY_CODE (or otherwise dropping a block from its tier) does NOT remove the now-excess SYNC block from its `.md` on disk — `process_file` only inserts missing blocks. Strip the excess block(s) from the agent `.md` source by hand (or a scoped one-off) so the regression suite's tier assertions pass.

**Skill-only blocks (orchestration).** A block in `SKILL_BLOCK_ORDER` but in NO agent tier reaches every skill and zero agents. TC-UAR-017 treats that shape as drift by default — a protocol that reached ≥3 skills but no agent is usually an oversight — so declare a genuinely skill-only block in BOTH lists:

- `.claude/hooks/tests/suites/agent-universal-rules.test.cjs` → `AGENT_ADOPTION_EXEMPT` — the ONLY list TC-UAR-017 reads. Omit the block here and the suite fails.
- `.claude/scripts/agent_protocol_matrix.py` → `EXCLUDED_ORCHESTRATION` — read only by that file's own `--validate` check (b) (`agent_protocol_matrix.py:481`), which rejects a per-agent manifest that assigns an excluded block without an `ORCHESTRATION_WHITELIST` entry. Omit it here and no test fails; the manifest is simply free to assign the block to an agent.

> **No cross-check exists.** The two lists are enforced by two SEPARATE mechanisms and nothing verifies they agree — TC-UAR-017 never reads `EXCLUDED_ORCHESTRATION` (it appears in that test file only in a comment and a failure-message string). Keeping them in step is a **convention**, not an enforced invariant: update both by hand in the same change.

The bar for that exemption is caller-side ownership: the block must drive orchestration the leaf either **cannot perform** (it has no access to the parent conversation, workflow state, or the user) or **has no basis to perform** (the decision was already made upstream before its brief was issued). Current members — `nested-task-creation`, `subagent-return-contract`, `sub-agent-selection`, `parallel-phase-advancement`, `parallel-subagent-dispatch`, `goal-contract-satisfaction-loop` — expand workflow steps, choose/brief sub-agents, partition a task list into parallel waves, or drive a user-facing convergence loop. `parallel-subagent-dispatch` is the second kind: an agent CAN spawn (agent files generally carry no `tools:` restriction), but it receives one already-partitioned brief, so re-running PAR/SEQ tagging inside it re-decides upstream's call. An agent that legitimately fans out carries that instruction in its own `.claude/agents/*.md` definition instead.

#### Step B3: Run the script (dry-run first)

```bash
python .claude/scripts/sync-hooks-to-skills.py --dry-run --verbose
# Verify: expected N updated, 0 errors

python .claude/scripts/sync-hooks-to-skills.py --verbose
# Verify: the computed on-disk tier inventory was processed; already-current targets skip
```

#### Step B4: Verify

```bash
# Confirm target files now contain the new block. Expected count is TIER-AWARE.
# The skill tier and the agent tiers are INDEPENDENT lists — a block reaches
# agents only if it is ALSO in an agent tier, so union the rows that apply:
#   - block in SKILL_BLOCK_ORDER ONLY     → every discovered skill, 0 agents
#                                           (skill-only ⇒ must be declared in BOTH
#                                            exemption lists — see "Skill-only blocks")
#   - block in SKILL_BLOCK_ORDER + CORE   → every discovered skill + every discovered agent
#   - block in CORE_BLOCK_ORDER           → every discovered agent (skills excluded)
#   - block in READONLY_CODE_BLOCK_ORDER  → 21 agents (17 code + 4 readonly-code)
#   - block in CODE_BLOCK_ORDER           → 17 code agents only
grep -rl "SYNC:new-block-name" .claude/skills/*/SKILL.md .claude/agents/*.md | wc -l

# Then run the agent-coverage regression suite — it asserts tier membership,
# disjointness, and SYNC tag balance across the discovered agent inventory.
node .claude/hooks/tests/run-all-tests.cjs --filter=agent-universal
```

Check a representative file of each affected tier manually to confirm placement and formatting.

---

## Usage Examples

```
$sync-skills-shared-protocols understand-code-first     # Sync one tag (Operation A)
$sync-skills-shared-protocols all                        # Sync all tags (Operation A)
$sync-skills-shared-protocols                            # Interactive — asks which tags
# Adding new block: use Operation B workflow above (script-driven)
```

## Rules

- ALWAYS edit `sync-inline-versions.md` FIRST, then run this skill
- NEVER modify content outside `<!-- SYNC:tag -->` boundaries
- NEVER touch `:reminder` blocks unless explicitly asked — and when asked, sync them WITH THE SCRIPT on the `{tag}:reminder` tag; never hand-write one
- If close tag is missing in a target file, SKIP that file and report it as an error
- Use the `Grep` tool (not shell grep) per project conventions
- Verify tag balance after every sync run
- For bulk-insert (Operation B): use `sync-hooks-to-skills.py` — NEVER do it manually across 288 files

---

<!-- SYNC:critical-thinking-mindset -->

> **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.

<!-- /SYNC:critical-thinking-mindset -->

<!-- SYNC:shared-protocol-duplication-policy -->

> **Shared Protocol Duplication Policy** — Inline protocol content in skills (wrapped in `<!-- SYNC:tag -->`) is INTENTIONAL duplication. Do NOT extract, deduplicate, or replace with file references. AI compliance drops significantly when protocols are behind file-read indirection. To update: edit `.claude/skills/shared/sync-inline-versions.md` first, then grep `SYNC:protocol-name` and update all occurrences.

<!-- /SYNC:shared-protocol-duplication-policy -->

<!-- SYNC:ai-mistake-prevention -->

> **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.

<!-- /SYNC:ai-mistake-prevention -->

<!-- SYNC:shared-protocol-duplication-policy:reminder -->

**IMPORTANT MUST ATTENTION** follow duplication policy: inline protocols are INTENTIONAL, never extract to file references

<!-- /SYNC:shared-protocol-duplication-policy:reminder -->

<!-- SYNC:critical-thinking-mindset:reminder -->

**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.

<!-- /SYNC:critical-thinking-mindset:reminder -->

<!-- SYNC:ai-mistake-prevention:reminder -->

**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.

<!-- /SYNC:ai-mistake-prevention:reminder -->

<!-- SYNC:project-protocol-overlay -->

> **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`.
>
> 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.

<!-- /SYNC:project-protocol-overlay -->

<!-- SYNC:project-protocol-overlay:reminder -->

**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.

<!-- /SYNC:project-protocol-overlay:reminder -->

## Closing Reminders

**IMPORTANT MUST ATTENTION Goal:** Two operations — (A) propagate updated content for existing SYNC: blocks across all skills, or (B) add a new SYNC: block to all skill/agent files that don't have it yet.

**IMPORTANT MUST ATTENTION Workflow:** Operation A: identify tag(s) → read canonical content → find targets → replace only block bodies → verify balance/parity/outside-block diff → handle reminders as separately requested; Operation B: edit canonical → update inserter tiers → dry-run → run → verify tier coverage and balance → report tags/files/errors.

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

- **Critical Thinking:** apply critical+sequential thinking; trace every claim, confidence >80%.
- **Shared Protocol Duplication:** inline protocols are intentional; never extract to file references.
- **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** edit `sync-inline-versions.md` FIRST before syncing to skills
**IMPORTANT MUST ATTENTION** verify SYNC tag balance after every sync run
**IMPORTANT MUST ATTENTION** NEVER modify content outside `<!-- SYNC:tag -->` boundaries
**IMPORTANT MUST ATTENTION** skip files with missing close tags and report as errors

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

<!-- CODEX:SYNC-PROMPT-PROTOCOLS:START -->
## Static 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` (legacy filename; static protocol composer)

## [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 immediately before the first target read, grep, edit, test, or analysis. 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. After compaction, resume, delegation, or a material context change, re-read the required docs and state `Reference docs read: ... | Not applicable: ...`; a hook reminder or prior conversation is not proof that the files are loaded. 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. **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.
7. **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.
- **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.
- **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 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.
- **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.
- **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.
- **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.

<!-- CODEX:SYNC-PROMPT-PROTOCOLS:END -->