cook
Pass
Audited by Gen Agent Trust Hub on Apr 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a rigorous development workflow including mandatory planning, research, and testing phases. It includes 'Rationalization Prevention' instructions to ensure the agent follows all safety and quality checkpoints.
- [COMMAND_EXECUTION]: The skill invokes a local script located at
.claude/scripts/code_graphto perform code analysis and dependency mapping. This is a standard part of the intended development tooling for codebase navigation. - [PROMPT_INJECTION]: The skill processes user-supplied arguments through a
<tasks>tag and reads local project documentation. While these represent ingestion points for untrusted data, the skill includes several 'hard-gates' and mandatory verification steps that mitigate the risk of malicious instructions influencing the agent's behavior.
Audit Metadata