contract-drafting

Fail

Audited by Gen Agent Trust Hub on Apr 7, 2026

Risk Level: HIGHCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill contains instructions for the agent to execute shell commands (ls) to inspect multiple hidden configuration directories such as ~/.claude/skills/ and ~/.agent/skills/. This allows the agent to fingerprint the user's environment and identify other installed tools without explicit user authorization.- [EXTERNAL_DOWNLOADS]: The skill directs the agent to offer the installation of additional 'companion skills' from an unverified third-party repository (AbsolutelySkilled/AbsolutelySkilled) using the npx command. Downloading and executing code from an unknown source poses a significant supply chain risk.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Apr 7, 2026, 03:04 PM
Security Audit — agent-trust-hub — contract-drafting