메뉴
Parse and classify enhancement specifications, then dispatch to the right specialist agents
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
The SOLE WRITER to the V3.0 mem0 second brain. Invoked agentically by the executive at write moments — post-run, post-retro, post-spec-merge, manual /harvest. Decides which facts to classify into which memory type (principle/semantic/procedural/episodic/reflective), validates against the schema, and calls the deterministic plumbing in references/ via Bash. Blocks until writes are durable (poll event endpoint for SUCCEEDED). Never invoked from worker code.
Consult the V3.0 mem0 second brain agentically during executive planning, work selection, pre-spawn memory pack building, and failure diagnosis. Use this skill whenever the executive needs to remember prior runs, retros, or learned principles. Read-only — never writes. Drives the unified `mem0` CLI via Bash, orchestrating multiple iterative search queries (mem0 does not do multi-hop graph traversal natively, so iterative search IS the multi-hop).
Hook D — Failure-Diagnosis memory consultation. Runs at executive Phase 7 (after 3+ consecutive failures on a work item) BEFORE the failure-diagnosis skill itself runs. Consults the second brain for similar prior failures and successful retry strategies, then hands enriched context to the failure-diagnosis skill. Read-only. Invokes the memory-reader skill.
Hook E — Post-Retrospective Harvest. Runs after a retrospective markdown doc completes (per-run or weekly). Reads the retro fully and decides which durable lessons (reflective, semantic, procedural memories) to write to the second brain. Typically 3–8 writes per weekly retro, 0–3 per per-run retro. Invokes the memory-harvester skill for the actual writes.
Hook C — Post-Run Harvest. Runs after a worker completes (executive Phase 6 success path). Decides agentically which facts from the run to write to the second brain — typically 0–3 memory writes per run (one episodic, optionally a semantic or procedural). Invokes the memory-harvester skill to do the actual writes (with event-polling for durability).
Hook B — Pre-Spawn Memory Pack build. Runs before each worker spawn (executive Phase 4 prep). Reads the second brain for project-specific memories and produces a Memory Pack markdown block that worker-spawner injects into the worker's generated CLAUDE.md. Workers consume this as static markdown — they never call mem0 themselves. Read-only. Invokes the memory-reader skill.
| name | enhance |
| description | Parse and classify enhancement specifications, then dispatch to the right specialist agents |
Parse enhancement specifications, classify each by type, and dispatch to the appropriate specialist agent. This skill is for the enhance orchestrator — it does NOT do implementation or testing directly.
{{TARGET_DIR}} — path to the generated study app{{HARNESS_ROOT}} — path to the harness repo{{ENHANCEMENT_SPECS}} — JSON array of enhancement spec objectsEach spec object has:
{
"id": "string — unique identifier",
"title": "string — human-readable title",
"priority": "number — lower = higher priority",
"description": "string — what this enhancement does",
"targets": [
{ "path": "relative/to/target", "action": "create|modify", "description": "what to do" }
],
"requirements": ["string — functional requirement"],
"references": ["path — files to read for context"],
"acceptance_criteria": ["string — how to verify"]
}
Classify each enhancement based on its targets paths:
| Target paths contain | Type | Delegate to |
|---|---|---|
src/components/, src/pages/, src/lib/, any .jsx/.tsx/.css file | ui | jack-web-build-v1 agent via Task (has Playwright MCP for browser testing) |
.claude/skills/, .claude/agents/, .md prompt files | prompt | Handle directly with Read/Write/Edit |
package.json, config files, build scripts | config | Handle directly with Bash |
If an enhancement has targets spanning multiple types, use the most capable agent — UI enhancements that also touch config should go to jack-web-build-v1.
Build a Task prompt that includes:
{{TARGET_DIR}}/ for app files, {{HARNESS_ROOT}}/ for harness files)The Task agent (jack-web-build-v1) already has Playwright MCP tools. Tell it to:
browser_evaluate if needed{{TARGET_DIR}}/ai-docs/phases/ENHANCE/changes-{id}.mdHandle directly — these don't need browser testing. Read, modify, verify.
cd {{TARGET_DIR}} && npm run build 2>&1{{TARGET_DIR}}/ai-docs/phases/ENHANCE/changes.mdWhen the enhancement involves code that calls the Claude API and expects JSON: Claude often wraps JSON in markdown fences (```json ... ```) even when told to return raw JSON. Any JSON.parse() on Claude API responses MUST strip fences first:
rawText = rawText.replace(/^```(?:json)?\s*\n?/i, '').replace(/\n?```\s*$/i, '').trim()
If the target project has deposited skills in .claude/skills/ (e.g., score-explanation, evaluate-rationale), the enhancement MUST read those skills and use their output schema — not invent a simplified one.
Reuse existing agents. The jack-web-build-v1 agent already knows how to build React apps and test with Playwright. Don't duplicate that capability — delegate to it.