自动记忆进化
v1.0.5Automatically summarize daily discussions and update memory files. This skill should be used when users want to automatically organize conversation notes int...
⭐ 0· 70·1 current·1 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
medium confidencePurpose & Capability
Name/description match the included Python scripts: both read/write files under ~/.openclaw/workspace, extract topics from markdown and update MEMORY.md. The scripts call the OpenClaw CLI to obtain session/chat history, which is appropriate for a memory summarizer.
Instruction Scope
SKILL.md instructs reading/writing only local workspace memory files and scheduling cron jobs. The code follows that scope, reading ~/.openclaw/workspace/{memory,data} and writing MEMORY.md and report files. It does run subprocesses (openclaw, python3, node) — expected for integrating with the OpenClaw CLI and optional node scripts — but the script paths referenced differ in places (see below).
Install Mechanism
Instruction-only skill with bundled Python scripts; no network downloads or external installers. No install spec present, so nothing is fetched or written automatically by the registry beyond the files included.
Credentials
The skill requests no environment variables or credentials. It operates on files under the user's OpenClaw workspace and invokes local CLIs (openclaw/node/python) which is proportionate to its function. No unrelated credentials or external endpoints are requested.
Persistence & Privilege
Skill writes persistent state and memory files under ~/.openclaw/workspace and is intended to be run regularly via cron/heartbeat. always:false (not force-enabled). Be aware that cron jobs and the heartbeat script will run in the background and create heartbeat-state.json and reports.
Assessment
This skill appears to do what it claims, but inspect and adjust before enabling persistent scheduling. Actions to consider:
- Confirm where you will place the files. SKILL.md cron paths point to ~/.openclaw/skills/auto-memory-evolution/scripts/*.py, but heartbeat-check.py expects daily-evolution.py under ~/.openclaw/workspace/scripts/ — fix or standardize the paths so heartbeat can find the script.
- Review and remove any optional node scripts or other files the heartbeat script may execute (skill-learning.js, model-health-check.js, daily-growth-report.js) if you don't want arbitrary Node code run from your workspace.
- Verify you have and trust the local OpenClaw CLI (the code calls subprocess ['openclaw', 'chat', 'history', ...]) and understand what data it returns; the script will read session/chat history returned by that CLI.
- Because the skill writes files and sets cron jobs, run it initially in a safe/test environment or with backups of ~/.openclaw/workspace to confirm behavior.
- If you need auditing, add logging or run the scripts manually first to observe output and ensure no unexpected network or file access occurs.Like a lobster shell, security has layers — review code before you run it.
automationvk97b5900p0vfd2w9hty322zpb183f2dflatestvk97c0etstkdjdn5zhm1s1135e583f99dmemoryvk97b5900p0vfd2w9hty322zpb183f2df
License
MIT-0
Free to use, modify, and redistribute. No attribution required.