메뉴
Use this skill when inspecting RECALL memory quality, when reviewing current memory state, when auditing noise or conflicts, or when gathering memory IDs before maintenance. Use proactively for inspection-only work, not mutation or new-memory capture.
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
Use this skill proactively when creating a reusable category, when configuring retrieval weight, when refining an auto-created definition, or when organizing repeated memory types. Trigger when category configuration is needed; invoke automatically for durable taxonomy work, not one-off tags.
Use this skill when a fresh RECALL session or new thread needs the usage contract, when the user asks what RECALL remembers, when prior project decisions or commands are relevant, or when a durable requirement, risk, decision, command, or project-state update needs to be saved. Auto-loads at sessionStart on Kimi Code; invoke before other RECALL skills so provider provenance and local-only storage rules are established.
Use this skill when changing RECALL capture mode, when cleaning noisy memory, when repairing local storage, or when updating lifecycle state after review. Use proactively for explicit maintenance requests, never for ordinary retrieval or new-memory capture.
Use this skill when the user explicitly asks to run, judge, or compare a RECALL benchmark (e.g. "run a judged normal bench", "recall-bench light", "benchmark RECALL against the baseline"). It runs the deterministic bench harness, optionally acts as the LLM judge for emitted judge tasks, aggregates scores, and reports results. Never invoke for ordinary RECALL memory usage, plugin development, or test-suite runs.
Use this skill proactively when debugging a known failure, when implementing work shaped by prior decisions, when testing with verified commands, or when starting a new thread that needs project state. Trigger when local memory prevents repeated investigation; invoke automatically only when project context is relevant.
Use this skill when deciding whether candidate context belongs in Recall memory, repo docs, skill/plugin instructions, provider config, or current chat only; when planning memory cleanup; when detecting stale, duplicate, conflicting, low-value, source-backed, command, or weak-preference memories; or when applying safe non-destructive hygiene. Use proactively for memory quality policy before using manage-memory for direct ID-based mutation.
| name | review-memory |
| description | Use this skill when inspecting RECALL memory quality, when reviewing current memory state, when auditing noise or conflicts, or when gathering memory IDs before maintenance. Use proactively for inspection-only work, not mutation or new-memory capture. |
Use this skill when the user asks what RECALL currently remembers, asks for a memory audit, or needs IDs before changing memory state.
RECALL is local-only project memory. Read from the active project's RECALL memory directory: .recall/ for new projects, or existing .codex_memory/ stores for legacy projects. Never require hosted services or external APIs. Do not repeat secrets verbatim if a stored memory appears to contain sensitive data.
Use this skill for:
Do not use this skill to create or mutate memory directly. It is an inspection and audit skill.
Use the public RECALL adapter:
Run these examples from the installed/source plugin root so ./scripts/recall_skill.py resolves. If the shell is in the active project repository, do not look for ./scripts there; use the adapter's absolute path and pass --root <project-root> when needed.
python ./scripts/recall_skill.py review-memory --limit 20
python ./scripts/recall_skill.py audit-memory --limit 20
Use recall_skill.py only. Treat lower-level backend scripts as internal support code.
This skill receives an inspection request and returns evidence: counts, IDs, quality signals,
and suggested follow-up. It does not mutate memory. If the user asks to edit, archive, confirm,
repair, or delete records, collect the needed IDs here and then hand the action to
manage-memory.
Use the contract asset as the quick boundary check:
{"asset":"assets/contract.json","kind":"inspection-boundary"}
This skill usually receives:
| Input shape | Use it for | Example |
|---|---|---|
| broad review request | default inspection | "what does recall currently know?" |
| quality or noise concern | audit mode | "is memory noisy?" |
| narrow slice request | filtered review | "show stale risks" |
| follow-up maintenance task | ID gathering | "which memories should I prune?" |
python ./scripts/recall_skill.py review-memory --status active --category requirements
python ./scripts/recall_skill.py review-memory --source finalizer --limit 50
python ./scripts/recall_skill.py review-memory --status stale --status superseded
review-memory --limit 20 when the request is broad.audit-memory --limit 20 when the user wants signal-vs-noise diagnosis, archive candidates, or noisy command patterns.--status, --category, or --source filters when the user asks about a specific slice.manage-memory commands such as confirm-memory, resolve-memory, stale-memory, supersede-memory, merge-memories, or prune-memory.Report counts and the most relevant IDs. Prefer concise summaries over dumping every record. If many low-value automatic memories are present, recommend archive-noise for non-destructive cleanup and reserve prune-memory for targeted archival by ID.
Return a compact review summary that includes:
{
"mode": "audit-memory",
"shown": 6,
"top_ids": [12, 18, 24],
"recommendation": "archive-noise"
}
Broad review:
python ./scripts/recall_skill.py review-memory --limit 20
Noise-focused audit:
python ./scripts/recall_skill.py audit-memory --limit 20
Focused stale memory slice:
python ./scripts/recall_skill.py review-memory --status stale --category risks --limit 20
Conflict-focused inspection:
python ./scripts/recall_skill.py list-conflicts
Quality audit before cleanup:
python ./scripts/recall_skill.py audit-memory --limit 50
| Request type | Preferred command | Why |
|---|---|---|
| general memory inventory | review-memory | balanced overview |
| signal-vs-noise diagnosis | audit-memory | highlights noisy patterns |
| filtered category slice | review-memory with filters | narrower evidence set |
| maintenance prep | review-memory first | collect IDs before changes |
--status, --category, or --source.archive-noise rather than deleting records outright.manage-memory once the needed IDs are known.See Manage Memory for cleanup, lifecycle changes, and repair work after inspection. See Retrieve Memory for query-driven lookup when the user already knows the topic. See Save Insight for creating a new durable memory when the review reveals something missing. See Audit signals for quality and noise indicators. Sibling routes: skills/manage-memory, skills/retrieve-memory, skills/save-insight.