agents-md-generator
Pass
Audited by Gen Agent Trust Hub on Mar 6, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute a provided Python script (
scripts/scaffold_agents_md.py) to analyze the repository. This script uses thesubprocessmodule to rungit rev-parse --show-toplevelfor determining the repository root path. - [PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it processes untrusted repository files and includes their contents in its output.
- Ingestion points: The script reads and parses content from local repository files, including
package.jsonscripts and GitHub Actions workflows (.github/workflows/). - Boundary markers: The generated Markdown documentation does not include explicit delimiters or warnings to isolate untrusted content from the agent's instructions.
- Capability inventory: The skill is designed to write files to the disk and suggest executable commands (e.g., install, test, build) for the agent to use.
- Sanitization: The tool extracts string values directly from the source configuration files and interpolates them into templates without sanitization or validation.
Audit Metadata