do-dox-install
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes existing repository files and documentation to construct new agent instruction sets (AGENTS.md), creating a surface where untrusted project data can influence the agent's logic.\n
- Ingestion points: The agent is instructed to read existing root AGENTS.md or CLAUDE.md files and scan the repository for folder-specific content to generate child documentation.\n
- Boundary markers: The instructions do not define boundary markers or explicit safety warnings to prevent the agent from obeying instructions that might be embedded within the data it is merging.\n
- Capability inventory: The skill requires the agent to perform file system scanning, file reading, and file writing operations.\n
- Sanitization: There is no evidence of sanitization or filtering of the source content before it is consolidated into the project's agent instructions.
Audit Metadata