ClawAudit verdict
hub1
The skill provides memory infrastructure for autonomous agents and does not exhibit malicious behavior.
⚠ 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 (6)
Possible hardcoded credential
scripts/sync-all-acp-agents.ts · prose · downgraded · API_KEY = 'acp-4e0e4e39028eda8e44a2
References child_process — can spawn system processes
scripts/full-reputation-sync.ts · prose · downgraded · child_process
Dynamic import() — loads module at runtime
src/app/api/reputation/index/route.ts · prose · downgraded · import('
Instructs covert action — may act without user awareness
src/lib/qdrant.ts · prose · downgraded · silently
Makes HTTP request to external URL
docs/AGENT_REPUTATION_SERVICE.md · code · fetch('https://
Accesses sensitive environment variables
src/app/api/auth/session/route.ts · prose · downgraded · process.env.JWT_SECRET
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class D). 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.