req-parser
Pass
Audited by Gen Agent Trust Hub on Feb 26, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface due to the processing of untrusted document content.\n
- Ingestion points: External documents (PDF, DOCX, etc.) located in the
original-requirements/directory.\n - Boundary markers: The content is delimited by horizontal rules and source metadata comments, which do not provide robust isolation from instructions embedded within the text.\n
- Capability inventory: The agent has access to
Bash,Read, andWritetools, which could be misused if a document hijacks the agent's instructions.\n - Sanitization: No filtering or sanitization of the extracted text is performed before it is merged and analyzed by the AI.\n- [EXTERNAL_DOWNLOADS]: Fetches Python dependencies and potential layout analysis models via the
uvtool and theDoclinglibrary from official package registries and well-known technology provider repositories.
Audit Metadata