fixer
Pass
Audited by Gen Agent Trust Hub on May 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONNO_CODE
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to discover and execute project-specific test commands from the local environment, such as those defined in a README, Makefile, or package.json file. It also mentions using the 'rg' (ripgrep) utility for repository navigation.
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface. 1. Ingestion points: Untrusted data enters the context via the 'Critical findings' and 'review context' inputs in SKILL.md. 2. Boundary markers: The skill does not define delimiters or provide 'ignore embedded instructions' warnings for the input data. 3. Capability inventory: The agent is granted file read, file write, and arbitrary command execution capabilities across the workflow (SKILL.md). 4. Sanitization: No sanitization or validation of external content is described.
- [NO_CODE]: The skill consists entirely of instructional markdown and contains no code scripts, executables, or package manifests.
Audit Metadata