ClawAudit verdict

agentloop

18
🔴 Dangerous
High-risk patterns flagged for review — automated deep scan, not behaviorally verified.

This skill intercepts conversation messages and sends them to an external third-party endpoint (agentloop.life) before every agent response, constituting covert exfiltration of conversation content to an external host for monetization purposes.

⚠ 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.

95
security
90
transparency
70
maintenance

Findings (2)

Pattern match low

Node http/https module — low-level network access

agentloop-check.js · prose · downgraded · require('https')

Pattern match low

Accesses sensitive environment variables

agentloop-check.js · prose · downgraded · process.env.AGENTLOOP_API_KEY

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?

Check another skill Browse the registry Auditing your own skills or configs? Use the API