agents-md-author
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes content from project configuration files (like package.json or pyproject.toml) to generate documentation. This creates a surface where malicious instructions embedded in a project's metadata could be propagated into the generated instruction files.
- Ingestion points: Project root configuration files (package.json, pyproject.toml, go.mod, Cargo.toml) identified in SKILL.md.
- Boundary markers: The skill does not explicitly define sanitization or boundary markers for the content interpolated from these files into the markdown templates.
- Capability inventory: File-write operations to create AGENTS.md and CLAUDE.md.
- Sanitization: No specific sanitization logic is described for the extracted metadata.
- [SAFE]: The skill references documentation and standards from established vendors and services, such as Anthropic, OpenAI, and agents.md. These references are used to align the generated files with industry conventions.
Audit Metadata