define-prioritization-framework
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied feature lists and descriptions to generate ranked reports for use in roadmap planning and launch checklists. \n
- Ingestion points: The skill accepts a list of candidate items, decision context, and customer research as primary inputs in SKILL.md. It also utilizes 'Read' and 'Grep' tools which can ingest additional local data. \n
- Boundary markers: The instructions lack explicit delimiters or 'ignore embedded instructions' warnings for the agent when processing the untrusted item descriptions. \n
- Capability inventory: The skill performs analytical synthesis and outputs Markdown artifacts that are intended to feed into downstream workflows like 'deliver-launch-checklist' and sprint planning. \n
- Sanitization: No sanitization, escaping, or schema validation is performed on the input strings before they are processed by the LLM for scoring.
Audit Metadata