Caveat verdict
zhihe-legal-research
Legal research tool connecting to a documented AI legal platform via shell scripts with a clean async task-polling workflow and no exfiltration behavior.
⚠ Flagged for review — coarse, uncorroborated signal, not a confirmed exploit. Review the config yourself before installing.
Automated static analysis — not a human review. Caveat flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.
Permission integrity
package_install
Findings (3)
Possible hardcoded credential
references/api-reference.md · code · TOKEN="your_token_here
Accesses OpenClaw config/secrets directly
scripts/token-manager.sh · prose · downgraded · ~/.openclaw/.env
Sets world-executable permissions
scripts/auth.sh · prose · downgraded · chmod 700
Permissions & capabilities
No declared permissions — minimal attack surface.
package_install Is this flag fair?
Thanks — recorded.