OpenClaw: memory freshness
v1.0.0为 OpenClaw 4.2 记忆搜索结果添加新鲜度警告。当记忆搜索结果返回时自动注入时间戳警告，提示模型验证旧记忆。当用户提到记忆过时、记忆不准确、记忆过期、或要求添加记忆新鲜度提醒时触发。
⭐ 0· 50·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The name/description (memory freshness warnings) align with the instructions: injecting rules into the system prompt, stamping memories with timestamps, and adding freshness annotations to search results are reasonable and expected for this purpose. No unrelated credentials, binaries, or network access are requested.
Instruction Scope
SKILL.md explicitly tells operators/agents to edit ~/.openclaw/openclaw.json, to add timestamps to memory files on write, and to run local verification commands (grep/rg) when using memory that points at files/functions. This is within the feature scope, but it does require the agent/user to perform file edits and local shell operations — ensure you understand the implications (see guidance).
Install Mechanism
Instruction-only skill with no install spec and no code files. Nothing is downloaded or written by an installer — lowest-risk install footprint.
Credentials
The skill declares no required environment variables, credentials, or config paths. The guidance to edit ~/.openclaw/openclaw.json is a behavioral recommendation rather than a hidden requirement, so requested access is proportional to the stated purpose.
Persistence & Privilege
always is false and disable-model-invocation is not set — normal. The skill does not request persistent elevated privileges or modifications to other skills' configurations. It suggests modifying the agent/system prompt (user action), which is expected for global behavior changes.
Assessment
This skill is coherent with its goal, but be aware of what it asks you to do: 1) It recommends editing your global OpenClaw system prompt (~/.openclaw/openclaw.json) — that change affects all agents, so back up the file before editing and review the new prompt for unintended instructions. 2) It suggests adding timestamps to memory files and running local commands (grep/rg) to verify memory contents; those verification steps will cause the agent or you to read local files referenced by memories—ensure you are comfortable with that access and that the referenced paths are safe to inspect. 3) The instructions assume grep or ripgrep (rg) are available; if not, install or adapt the verification commands. If you want stricter safety, apply the changes in a test environment first and limit which memories or agents use the new prompt.Like a lobster shell, security has layers — review code before you run it.
latestvk97eman8kq2gw71s42htmcebgn847thm
License
MIT-0
Free to use, modify, and redistribute. No attribution required.