ClawAudit verdict

openclaw-computer-use

openclaw-computer-use-v2-0-0-tar

88
๐ŸŸข Trusted
Low risk โ€” reviewed by ClawAudit, behavior matches stated purpose

GUI automation skill enabling desktop control for clearly stated automation purposes; package installs are documented dependencies and credential_store is for session persistence, all matching stated purpose.

โš  Flagged for review โ€” coarse, uncorroborated signal, not a confirmed exploit. Review the config yourself before installing.

Automated static analysis โ€” not a human review. ClawAudit flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.

43
security
70
transparency
70
maintenance

Permission integrity

Installs packages at runtime โ€” transitive dependencies are not auditable

package_install

Findings (4)

Pattern match high

References sudo โ€” requests elevated privileges

SKILL.md ยท code ยท sudo

Pattern match high

apt-get install โ€” installs system packages

SKILL.md ยท code ยท apt-get install

Pattern match medium

subprocess execution โ€” runs system commands from Python

computer_use_pro.py ยท prose ยท downgraded ยท subprocess.run(

Pattern match low

Python shutil file operation โ€” copies/moves/deletes files

computer_use_pro.py ยท prose ยท downgraded ยท shutil.copytree(

Why the tier is capped

Execution sink present in raw bytes (Hard Floor: class B/D). Final tier capped at Caution โ€” cannot be lifted by any downgrade, example-payload opt-in, or allowlist.

Permissions & capabilities

No declared permissions โ€” minimal attack surface.

package_installcredential_store

Is this flag fair?

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