ClawAudit verdict
dreamlover-skill
The skill is designed to manage intake information for character skills, with no indication of malicious intent.
⚠ 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.
Findings (3)
Writes to SKILL.md — self-modifying skill
tools/skill_writer.py · prose · downgraded · SKILL.md").write
Instructs covert action — may act without user awareness
prompts/style_examples_builder.md · prose · downgraded · secretly
Python shutil file operation — copies/moves/deletes files
tools/version_manager.py · prose · downgraded · shutil.copytree(
Permissions & capabilities
No declared permissions — minimal attack surface.
Is this flag fair?
Thanks — recorded.