ClawAudit verdict
interclaw
PGP-signed agent-to-agent email coordination skill using standard email infrastructure; all communication is explicit and user-directed with no hidden exfiltration.
⚠ 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 prompt injection — attempts to redefine agent identity
SKILL.md · prose · downgraded · You are now
Popular HTTP library — network access
README.md · code · Got
Instructs covert action — may act without user awareness
docs/protocol-v3.md · prose · downgraded · silently
Long base64 string (100+ chars) — likely obfuscated payload
README.md · prose · downgraded · PHN2ZyB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciIHZpZXdCb3g9IjAgMCAyNCAyNCI+
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
Requires 10 environment variables. (1 sensitive: PGP_PRIVATE_KEY_ID). Requires 1 system binary.
Is this flag fair?
Thanks — recorded.