ClawAudit verdict
safeagentdb
The skill focuses on setting up and maintaining SafeAgentDB-style database safety infrastructure without demonstrating any malicious or deceptive 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 (3)
References agent configuration files
references/agent-packaging.md · code · CLAUDE.md
References child_process — can spawn system processes
templates/scripts/supabase/apply-migrations.ts · prose · downgraded · child_process
Accesses sensitive environment variables
templates/scripts/supabase/cleanup-orphan-preview-branches.ts · prose · downgraded · process.env.VERCEL_TOKEN
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class B/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.