architecture
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill consists entirely of markdown instructions and YAML frontmatter. It does not include any executable scripts (Python, Node.js, Shell, etc.), binary files, or external dependencies.
- [PROMPT_INJECTION]: The instructions are clear and instructional without attempting to override the agent's underlying safety guardrails or system prompts. Phrases like 'Important' and 'Do not' are used legitimately to define the task boundaries (e.g., 'Do not propose future behavior').
- [EXTERNAL_DOWNLOADS]: There are no network operations, URL references, or external resource fetches defined in the skill.
- [COMMAND_EXECUTION]: While the skill's purpose is to guide an agent in analyzing code (which may involve the agent using its own internal tools to read files), the skill itself does not specify any shell commands or executable code blocks to be run by the system.
- [DATA_EXFILTRATION]: No data exfiltration patterns were detected. The skill focuses on generating local documentation files.
Audit Metadata