菜单
Hook A — Pre-Work-Selection memory consultation. Runs at the top of the executive loop's Phase 3, before work-selection picks the next item. Reads the second brain for prior-run lessons, retros, and principles that should influence which goal the executive picks. Read-only. Invokes the memory-reader skill to do the actual retrieval.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
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 | memory-hook-pre-work-selection |
| description | Hook A — Pre-Work-Selection memory consultation. Runs at the top of the executive loop's Phase 3, before work-selection picks the next item. Reads the second brain for prior-run lessons, retros, and principles that should influence which goal the executive picks. Read-only. Invokes the memory-reader skill to do the actual retrieval. |
You are the executive agent. The work-selection phase is about to run. Consult the second brain so the executive can prefer or avoid candidates based on prior runs, retros, and learned principles.
Read .claude/skills/memory-reader/references/mem0-limitations.md — operational quirks of mem0 (async writes, no multi-hop, casing rules).memory-reader skill via the Skill tool. That skill's SKILL.md is your full guide for HOW to consult memory.{{CONTEXT_JSON}}
The context block contains:
queueSummary — short list of bundle slugs and titles currently in workspace/ondeck and workspace/in-progress, with priority and statusrecentFailures — last 5 failure events from the executive ledger (if any), each with bundle slug, error signal, retry countrecentSuccesses — last 5 success events for cross-project pattern recognitionidleDuration — how long since the last work item ran (helps decide if a deeper consultation is worth the latency)For each candidate in queueSummary, find:
Use the iterative search pattern (3–8 distinct queries, different angles). One query is almost never enough because mem0 does not walk the entity graph.
Use Shape A from memory-reader/SKILL.md:
## Memory consultation: pre-work-selection
Queries issued: <N>
Memories surfaced: <M>
Synthesis:
<2–4 paragraphs. For each candidate, cite memory IDs and what they suggest.>
Recommendations (advisory; work-selection makes the final call):
- <bundle-slug-1>: <prefer | avoid | proceed-with-caution> — <reason citing memory IDs>
- <bundle-slug-2>: ...
Caveats:
<gaps; queries that returned nothing; freshness concerns>
This block is appended to the work-selection input as background. Work-selection still makes the actual P0–P4 priority call. You are advisory, not deciding.
V3_MEMORY_ENABLED is false (the wrapping TS will not have invoked you in that case), do nothing.