check
Pass
Audited by Gen Agent Trust Hub on Apr 5, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: Executes local commands for git operations and project testing (
git diff,npm test,cargo check, etc.) as part of its core review and verification functionality. - [EXTERNAL_DOWNLOADS]: Includes
git fetch originto synchronize local repository state with the remote origin. - [PROMPT_INJECTION]: The skill processes untrustworthy data from git diffs and task files while possessing file modification and command execution capabilities, creating a surface for indirect prompt injection.
- Ingestion points: Reads content from
git diffand local task files (SKILL.md). - Boundary markers: No explicit markers are defined to isolate the diff content from the agent's instructions.
- Capability inventory: Possesses
Bash,Edit,Write, andAgenttools across its scripts. - Sanitization: Does not perform validation or escaping of the input data before processing.
Audit Metadata