wiki-agents-md
Pass
Audited by Gen Agent Trust Hub on Feb 27, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: Indirect Prompt Injection Surface detected in the documentation generation logic.
- Ingestion points: The skill reads various configuration files including package.json, Makefile, pyproject.toml, and source code files from the user's repository (Step 3 and Step 4).
- Boundary markers: Absent. The generated AGENTS.md files do not use delimiters or explicit 'ignore' warnings to isolate external content from the instructions meant for the agent.
- Capability inventory: The skill performs standard filesystem operations such as listing directories (ls) and reading files.
- Sanitization: Absent. The skill instructions do not specify any validation or escaping of the content extracted from project files before it is included in the final markdown output.
Audit Metadata