skills/amit-t/skills/prd-review-panel/Gen Agent Trust Hub

prd-review-panel

Pass

Audited by Gen Agent Trust Hub on Oct 24, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external PRD files and incorporates their content into prompts for sub-agents. This is an inherent surface for indirect prompt injection given the tool's purpose of document analysis.
  • Ingestion points: Reads PRD files from outputs/prds/ and context-library/prds/, as well as strategy/research context from context-library/.
  • Boundary markers: Employs labeled sections (e.g., "PRD Content:", "Strategic Context:") in sub-agent prompts, but does not use explicit delimiters or instructions to ignore potential commands within the PRD text.
  • Capability inventory: Performs file read/write operations within the project directory structure and spawns multiple sub-tasks via parallel tool calls.
  • Sanitization: The skill does not perform pre-processing or sanitization of the PRD content before passing it to the sub-agent personas.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 24, 2026, 09:36 PM