agent-md-refactor
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides instructions for the agent to ingest and restructure user-controlled documentation files (such as
CLAUDE.md), which represents a surface for indirect prompt injection. - Ingestion points: The agent reads the content of existing instruction files found in the project root.
- Boundary markers: There are no explicit delimiters or instructions to ignore embedded prompts within the files being refactored.
- Capability inventory: The skill's capabilities are restricted to reading text files and writing new Markdown files to the project directory.
- Sanitization: The skill does not perform specific sanitization of the input text, as its primary purpose is the structural reorganization of that content.
- [NO_CODE]: The skill consists entirely of Markdown instructions, templates, and process descriptions. It does not include any Python scripts, Node.js code, shell scripts, or binary executables.
Audit Metadata