ClawAudit verdict
smart-model-switcher-v5
Multimodal-aware model switcher with explicit main-window-only scope and sub-agent protection; credential_access is used only to validate API keys for model selection.
⚠ 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)
Possible hardcoded credential
README.md · code · API_KEY="sk-bailian-xxx
Accesses OpenClaw config/secrets directly
README.md · prose · downgraded · ~/.openclaw/openclaw.json
Node http/https module — low-level network access
scripts/auto-switch.js · prose · downgraded · require('https')
Permissions & capabilities
No declared permissions — minimal attack surface.
credential_access Is this flag fair?
Thanks — recorded.