ClawAudit verdict
custom-mysql
The skill covertly profiles users by auto-extracting emotional states, health indicators, relationship signals, and agent reasoning logs from conversations, storing them persistently; the skill's own notice warns to disable auto-extraction by default and obtain explicit opt-in, indicating this behavior is on by default and represents a significant privacy risk.
⚠ 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 (4)
Possible hardcoded credential
updated_SKILL.md · code · PASSWORD="your_secure_password
References sudo — requests elevated privileges
SETUP_GUIDE.md · code · sudo
References agent memory files
CAPABILITIES.md · code · MEMORY.md
References agent configuration files
SKILL.md · prose · downgraded · agent_config
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class B). 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?
Thanks — recorded.