init-project-light
Pass
Audited by Gen Agent Trust Hub on Aug 24, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to read and analyze existing files in a directory to infer project context. This behavior introduces a surface for indirect prompt injection, as malicious instructions embedded in those files could theoretically influence the agent's behavior during the initialization process.
- Ingestion points: All text files with extensions .md, .tex, .bib, and .txt in the current directory (Phase 1, step 2).
- Boundary markers: None specified to differentiate between data and instructions.
- Capability inventory: File system operations including mkdir, touch, Read, Write, and Edit (SKILL.md frontmatter).
- Sanitization: No content filtering or validation is implemented for the data read from external files.
Audit Metadata