ClawAudit verdict
skill-publisher
skill-publisher-universal
The skill provides a generic publishing workflow for Claude Code skills. No malicious behavior detected.
โ Flagged for review โ coarse, uncorroborated signal, not a confirmed exploit. Review the config yourself before installing.
Automated static analysis โ not a human review. ClawAudit flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.
Permission integrity
network_out
Findings (2)
HTTP request to bare IP address โ common in malicious payloads
SKILL.md ยท code ยท http://192.168.28.92
Python shutil file operation โ copies/moves/deletes files
scripts/generate_package.py ยท prose ยท downgraded ยท shutil.copytree(
Permissions & capabilities
No declared permissions โ minimal attack surface.
network_out Is this flag fair?
Thanks โ recorded.