Self-Improvement Plus
v2.0.0Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
⭐ 0· 55·0 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 (capture learnings, errors, corrections) aligns with the instructions: create .learnings/ files, append entries, and optionally promote learnings to workspace docs. The only external resource referenced for manual install is a public GitHub repo, which is coherent with the described OpenClaw integration.
Instruction Scope
SKILL.md instructs the agent to create and update .learnings/ files in the project or ~/.openclaw/workspace and to promote learnings to workspace files or share across sessions. It also references inter-session tools (sessions_history, sessions_send). The doc explicitly warns not to log secrets and to sanitize summaries, but much of the sanitization is left to the agent/installer — this creates a data-exfiltration risk if users or the agent log raw transcripts or secrets when promoting or sending learnings.
Install Mechanism
This is instruction-only (no install spec). The doc mentions an optional manual git clone from a GitHub repo and copying hooks into ~/.openclaw/hooks. Those are explicit manual steps and use a well-known host; there is no automatic download/install declared by the registry entry.
Credentials
The skill requires no environment variables, credentials, or config paths beyond writing files into the project or the user's OpenClaw workspace (~/.openclaw/workspace and ~/.openclaw/hooks if the user opts in). File-system writes are proportional to the stated logging purpose, but these writes can expose sensitive info if not sanitized.
Persistence & Privilege
always:false (no forced inclusion). The skill documents an optional 'hook' that, if installed and enabled by the user, will run automatically at session start — that increases persistence. Because enabling hooks requires an explicit action (copy + openclaw hooks enable), this is an opt-in risk; users should inspect hook code before enabling.
Assessment
This skill appears to do what it says: create and maintain local .learnings files and optionally integrate with OpenClaw. The main risks are accidental data leakage when promoting or sharing learnings or when enabling hooks that run automatically. Before installing or enabling the hook: (1) inspect any hook scripts you copy from the GitHub repo to ensure they don't exfiltrate data, (2) avoid logging raw transcripts, secrets, tokens, or full config files — follow the SKILL.md advice to redact/summarize, and (3) only enable cross-session sharing (sessions_send, sessions_history) in trusted environments. If you plan to use the manual git clone or copy hooks, review that repository contents locally before enabling any automated behavior.Like a lobster shell, security has layers — review code before you run it.
errorsvk97d5f1bkmq5ke2bresfeyabnh846dvqimprovementvk97d5f1bkmq5ke2bresfeyabnh846dvqlatestvk97d5f1bkmq5ke2bresfeyabnh846dvqlearningvk97d5f1bkmq5ke2bresfeyabnh846dvq
License
MIT-0
Free to use, modify, and redistribute. No attribution required.