ClawAudit verdict

agent-desktop

aoto

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

Desktop GUI automation CLI wrapper using native OS accessibility APIs; the agent-desktop tool is explicitly documented as a local tool that outputs structured JSON for the agent to interpret, with no network exfiltration.

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.

93
security
70
transparency
90
maintenance

Permission integrity

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

package_install

Findings (1)

Pattern match low

Base64 encoding/decoding

references/commands-observation.md ยท prose ยท downgraded ยท base64-encode

Permissions & capabilities

No declared permissions โ€” minimal attack surface.

package_install

Is this flag fair?

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