writing-prds
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [SAFE]: No security issues were identified. The skill consists of instructional text and references to external product management resources.
- [INDIRECT_PROMPT_INJECTION]: The skill guides the transformation of untrusted user input into project specifications. 1. Ingestion points: Abstract user ideas and context provided for PRD drafting (SKILL.md). 2. Boundary markers: Absent. 3. Capability inventory: None; the skill does not request or use shell tools, network utilities, or file-writing capabilities. 4. Sanitization: Absent. Given the lack of dangerous capabilities, this represents a benign instructional surface.
- [EXTERNAL_DOWNLOADS]: The skill provides links to external templates and reference materials on well-known platforms such as Google Docs, Coda, and YouTube. These are for user reference and are not executed by the agent. Evidence: Links to PM templates and guest insights (references/artifacts.md, references/guest-insights.md).
Audit Metadata