self-improving-agent
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
v3.0.13
2026/04/03
re-upload no changes
v3.0.12
2026/04/02
No changelog provided.
v3.0.8
2026/03/27
No changelog provided.
v3.0.5
2026/03/17
No changelog provided.
v3.0.2
2026/03/14
No changelog provided.
v3.0.1
2026/03/11
No changelog provided.
v1.0.11
2026/02/22
No changelog provided.
v1.0.7
2026/02/21
No changelog provided.
v1.0.5
2026/02/03
No changelog provided.
v1.0.4
2026/01/31
No changelog provided.
v1.0.2
2026/01/26
No changelog provided.
v1.0.1
2026/01/19
No changelog provided.
v1.0.0
2026/01/05
No changelog provided.
Security Scan
Status
clean
OpenClaw
gpt-5-mini
OpenClaw analysis
The skill's code, hooks, and instructions are coherent with its stated purpose of capturing learnings and reminders; it doesn't ask for unrelated credentials or download arbitrary remote code, but you should review and opt-in to the hooks before enabling them.
Confidence: high
VirusTotal
Type: OpenClaw Skill Name: self-improving-agent Version: 3.0.13 The 'self-improving-agent' skill bundle is designed to help AI agents track errors, corrections, and new insights by logging them to local markdown files. The bundle includes shell scripts (e.g., `extract-skill.sh`, `error-detector.sh`) and OpenClaw hooks (`handler.js`) that facilitate this workflow by providing reminders and scaffolding new skill directories. The instructions in `SKILL.md` explicitly advise against logging secrets or sensitive data, and the provided scripts contain basic path validation to prevent directory traversal. No evidence of data exfiltration, malicious execution, or unauthorized persistence was found.