Caveat verdict

research-queue

88
🟢 Trusted
No high-risk patterns surfaced by the deep scan — automated capability review, not behavioral proof.

Manages a local QUESTIONS.md queue file and investigates open questions using web search or bounded local experiments; all operations are local with explicit rules against modifying production code without user request.

⚠ 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.

50
security
90
transparency
70
maintenance

Findings (2)

Pattern match critical

Writes to SKILL.md — self-modifying skill

references/automation.md · code · SKILL.md` and `QUESTIONS.md`. If the queue is missing, initialize it per the ski

Pattern match high

Accesses OpenClaw config/secrets directly

references/queue-format.md · code · ~/.openclaw/openclaw.json

Permissions & capabilities

No declared permissions — minimal attack surface.

Is this flag fair?

Check another skill Browse the registry Auditing your own skills or configs? Use the API