ClawAudit verdict
model-router-manager
The model-router-manager skill seems to be used for intelligent multi-model routing. It does not appear to have any 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 (2)
apt-get install — installs system packages
scripts/model-router.sh · prose · downgraded · apt-get install
Accesses OpenClaw config/secrets directly
scripts/model-router.sh · prose · downgraded · ~/.openclaw/openclaw.json
Permissions & capabilities
Requires 1 environment variable. Requires 1 system binary.
Is this flag fair?
Thanks — recorded.