ClawAudit verdict

improvement-learner

auto-improvement-learner

88
🟢 Trusted
Low risk — reviewed by ClawAudit, behavior matches stated purpose

Skill quality evaluator that scores SKILL.md files across 6 dimensions (accuracy, coverage, reliability, efficiency, security, trigger_quality) using local Python scripts with a HOT/WARM/COLD memory system; all operations are local.

⚠ Flagged for review — coarse, uncorroborated signal, not a confirmed exploit. Review the config yourself before installing.

Automated static analysis — not a human review. ClawAudit flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.

48
security
90
transparency
70
maintenance

Findings (5)

Pattern match high

Writes to SKILL.md — self-modifying skill

tests/test_learner.py · prose · downgraded · SKILL.md").write

Pattern match medium

Uses exec() — may execute shell commands

SKILL.md · prose · downgraded · exec(

Pattern match medium

os.system/popen — direct OS command execution

SKILL.md · prose · downgraded · os.system(

Pattern match medium

subprocess execution — runs system commands from Python

scripts/self_improve.py · prose · downgraded · subprocess.run(

Pattern match low

Python shutil file operation — copies/moves/deletes files

scripts/self_improve.py · prose · downgraded · shutil.copytree(

Why the tier is capped

Execution sink present in raw bytes (Hard Floor: class D). Final tier capped at Caution — cannot be lifted by any downgrade, example-payload opt-in, or allowlist.

Permissions & capabilities

No declared permissions — minimal attack surface.

Is this flag fair?

Check another skill Browse the registry Auditing your own skills or configs? Use the API