ClawAudit verdict
manusilized
This skill enhances OpenClaw with features like incremental streaming and tool-call fallback, no apparent 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 (3)
HTTP request to bare IP address — common in malicious payloads
patches/ollama-models.ts · prose · downgraded · http://192.168.20.14
Uses exec() — may execute shell commands
patches/ollama-stream.ts · prose · downgraded · exec(
queueMicrotask — defers execution via microtask queue
patches/ollama-stream.ts · prose · downgraded · queueMicrotask(
Permissions & capabilities
No declared permissions — minimal attack surface.
Is this flag fair?
Thanks — recorded.