ClawAudit verdict
agentfarm-finder
Twitter/social monitoring tool for AI Agent and crypto projects with keyword-based search; runs a local bash script on a cron schedule and writes CSV output files locally — no exfiltration, no credential access, no remote code execution despite executionSinkDetected flag (the bash invocation is a documented local script).
⚠ 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)
Unicode homoglyph detected — uses lookalike characters to evade pattern matching
output/results_2026-03-12_filtered.csv · prose
subprocess execution — runs system commands from Python
agentfarm.sh · prose · downgraded · subprocess.run(
References sudo — requests elevated privileges
output/results_2026-03-20_filtered.csv · prose · downgraded · sudo
Popular HTTP library — network access
output/results_2026-03-12.md · prose · downgraded · Got
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.