pipeline-manifest
Pass
Audited by Gen Agent Trust Hub on Aug 24, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted external data from project files which could potentially contain malicious instructions intended to influence the agent's behavior during the manifest generation or header insertion process.
- Ingestion points: The skill reads content from research scripts (.py, .R, .do, .jl, .m) and LaTeX files (.tex) during Phase 2 and Phase 4.
- Boundary markers: The instructions do not specify delimiters or safety warnings to distinguish between file content and agent instructions.
- Capability inventory: The agent uses Read, Write, Edit, Glob, and Grep to analyze and modify files based on extracted data.
- Sanitization: No explicit sanitization or validation of the extracted file content is performed before interpolation into pipeline.md or script headers.
Audit Metadata