project-coding-companion
Pass
Audited by Gen Agent Trust Hub on Mar 11, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted user prompts and local repository files during its evaluation and understanding phases. This creates a surface for indirect prompt injection where malicious instructions embedded in a codebase could influence agent behavior. Ingestion points: user prompts and repository files. Boundary markers: absent. Capability inventory: file modification and Bash execution. Sanitization: none mentioned.
- [COMMAND_EXECUTION]: The skill identifies the need to run Bash commands for development tasks like installs and tests. It includes safety checks, requiring user confirmation for destructive actions and avoiding automated git commits.
Audit Metadata