菜单
Use when reading from or writing to memory and you need guardrails against stale records, bloated memory files, or low-value writes that will not remain useful after the current workstream
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
Use when verifying implementation work and the failure mode is superficial approval, code-reading in place of execution, or over-trusting a passing test suite without trying to break the change
Use when coordinating multi-step work across workers or subagents, especially when research, implementation, and verification should be split while the main agent stays responsible for synthesis and user communication
Use when the task is codebase exploration rather than modification, and you need a fast, read-only search pattern that starts broad when location is unknown, narrows when location is known, and parallelizes independent searches
Use when deciding what information belongs in memory, how to classify it, and what must not be saved because it is derivable from the repo, git history, or current task context
Use when deciding whether shell commands are truly read-only, safe for concurrent execution, and free of path, git, redirection, or command-shape risks that should force a stricter permission path
Use when the real risk is not lack of knowledge but self-justifying shortcuts such as explaining instead of running commands, skipping the hard part first, or deciding on your own that verification would take too long
| name | smart-memory-guard |
| description | Use when reading from or writing to memory and you need guardrails against stale records, bloated memory files, or low-value writes that will not remain useful after the current workstream |
Memory is useful only while it stays accurate, compact, and worth loading. This skill operationalizes the guardrails implied by src/memdir/memoryTypes.ts, src/services/SessionMemory/prompts.ts, and src/services/extractMemories/prompts.ts.
Treat memory as a snapshot, not as present truth.
Before acting on a memory:
If memory conflicts with the repo or source of truth, trust what you observe now and update or remove the stale memory.
Oversized memory degrades prompt quality.
When a memory file starts crowding the prompt:
Do not let memory become a second transcript.
Before saving anything, ask:
If the answer is no, do not save it.
Reject writes like:
Those are activity logs, not durable memory.
Accept writes like:
You have applied this skill correctly when memory is verified before use, compressed before it becomes noise, and written only when the information is durable enough to earn a place in future context.
