Caveat verdict

gui-agent

gui-claw

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

GUI automation skill using screenshot-based detection; all operations are local with no credential exfiltration, behavior matches the stated automation purpose.

⚠ 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.

0
security
90
transparency
70
maintenance

Findings (9)

Pattern match high

Pipe to python — executes piped content as Python code

benchmarks/osworld/multi_apps.md · prose · downgraded · | python

Pattern match high

HTTP request to bare IP address — common in malicious payloads

DESIGN_unified_actions.md · code · http://172.16.105.128

Pattern match high

References sudo — requests elevated privileges

skills/gui-setup/SKILL.md · code · sudo

Pattern match high

subprocess execution — runs system commands from Python

platforms/macos.md · code · subprocess.run(

Pattern match medium

Instructs covert action — may act without user awareness

actions/_actions_macos.yaml · prose · downgraded · silently

Pattern match medium

Accesses OpenClaw config/secrets directly

docs/README_CN.md · prose · downgraded · ~/.openclaw/openclaw.json

Pattern match medium

subprocess with shell=True — command injection vector

scripts/gui_agent.py · prose · downgraded · subprocess.run(cmd, shell=True

Pattern match low

Python os.environ.get — reads environment variable

platforms/detect.py · prose · downgraded · os.environ.get(

Pattern match low

Python shutil file operation — copies/moves/deletes files

scripts/agent.py · prose · downgraded · shutil.copy(

Why the tier is capped

Execution sink present in raw bytes (Hard Floor: class B/D). 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?

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