crit
Pass
Audited by Gen Agent Trust Hub on May 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes the 'crit' and 'gh' binaries via the shell to manage code reviews, local browser-based feedback loops, and GitHub Pull Request synchronization.
- [PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection through its processing of the 'review.json' file.
- Ingestion points: The agent reads and acts upon structured comments from 'review.json' (Step 3).
- Boundary markers: Absent; there are no instructions to the agent to treat this data as untrusted or to ignore instructions embedded within the comments.
- Capability inventory: The skill can modify source code files and execute CLI tools ('crit', 'gh').
- Sanitization: Absent; the skill does not specify any validation or sanitization of the JSON content before it is used to influence the agent's code editing and command execution tasks.
Audit Metadata