ClawAudit verdict
skill-creator1
skill-vetter1
The skill is designed for skill creation and does not contain malicious code.
β Flagged for review β coarse, uncorroborated signal, not a confirmed exploit. Review the config yourself before installing.
Automated static analysis β not a human review. ClawAudit flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.
Findings (4)
Writes to SKILL.md β self-modifying skill
skill-creator/ζ΅θ―δΈζ/test_package_skill.py Β· prose Β· downgraded Β· SKILL.md").write
Attempts to modify skills or system prompts
skill-creator/ζ΅θ―δΈζ/ζ΅θ―δΈζ.py Β· prose Β· downgraded Β· Edit SKILL.md
Popular HTTP library β network access
skill-creator/ζ΅θ―δΈζ/quick_validate.py Β· prose Β· downgraded Β· got
Python shutil file operation β copies/moves/deletes files
skill-creator/ζ΅θ―δΈζ/test_package_skill.py Β· prose Β· downgraded Β· shutil.rmtree(
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class F). Final tier capped at Caution β cannot be lifted by any downgrade, example-payload opt-in, or allowlist.
Permissions & capabilities
No declared permissions β minimal attack surface.
Is this flag fair?
Thanks β recorded.