review-proposals
Pass
Audited by Gen Agent Trust Hub on May 15, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill involves processing proposals which could theoretically contain indirect instructions (Indirect Prompt Injection). However, the skill implements a mandatory human-in-the-loop validation process. It requires the agent to display the full proposal and a diff of the changes, and specifically wait for explicit confirmation from a user before applying any updates to the configuration file.
- [COMMAND_EXECUTION]: The skill modifies a configuration file located at
~/.claude/plugins/config/claude-for-legal/commercial-legal/CLAUDE.md. This modification is constrained to a specific, non-sensitive application configuration path and is the primary stated purpose of the skill.
Audit Metadata