agent-md-refactor
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-provided agent instruction files (e.g.,
CLAUDE.md,AGENTS.md), which represents a potential attack surface for indirect prompt injection if those files contain malicious instructions. - Ingestion points: Reads content from existing documentation files in the project root as described in the Refactoring Process.
- Boundary markers: The skill employs a structured 5-phase process and specific templates for the output files, providing organizational boundaries for the processed content.
- Capability inventory: The skill's capabilities are limited to reading, analyzing, and writing markdown files; it does not have access to network tools, shell execution, or sensitive environment variables.
- Sanitization: The skill organizes and prunes content (Phase 5) but does not explicitly sanitize the text of the instructions for potential injection attacks beyond identifying contradictions.
Audit Metadata