Caveat verdict
skill-expert-skills
skill-expert-skills-openclaw
Six-phase workflow tool for creating and optimizing SKILL.md files operating on local files with no network or credential access.
⚠ 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.
Findings (6)
Attempts to modify skills or system prompts
SKILL.md · code · Modify SKILL.md
Pipe to python — executes piped content as Python code
references/hypothesis-ladder-for-skills.md · code · | Python
Uses eval() — can execute arbitrary code
references/skill-discovery-protocol.md · prose · downgraded · eval(
Pipe-to-shell pattern (curl | sh) — supply chain attack vector
references/skill-discovery-protocol.md · prose · downgraded · curl | sh
Pipe to sh — executes piped content as shell commands
references/skill-discovery-protocol.md · prose · downgraded · | sh
Popular HTTP library — network access
scripts/quick_validate.py · prose · downgraded · got
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class A/B/F). Final tier capped at Caution — cannot be lifted by any downgrade, example-payload opt-in, or allowlist.
Permissions & capabilities
No declared permissions — minimal attack surface.
Is this flag fair?
Thanks — recorded.