ClawAudit verdict
healthfit
healthfit-cn
The skill provides a comprehensive health management system, involving motion training plans, dietary suggestions, and health data tracking. The capabilities are extensive and could be risky if not properly managed.
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)
Pipe to sh β executes piped content as shell commands
data/txt/glossary_tcm.txt Β· prose Β· downgraded Β· | sh
References agent memory files
HealthFit_Skill_ζε»Ίζ₯ε.md Β· code Β· daily_note
References agent configuration files
AGENTS.md Β· prose Β· downgraded Β· CLAUDE.md
Python shutil file operation β copies/moves/deletes files
scripts/backup.py Β· prose Β· downgraded Β· shutil.copytree(
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class A). 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.