prd-to-plan
Pass
Audited by Gen Agent Trust Hub on Aug 26, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from user-provided PRDs to generate implementation plans.
- Ingestion points: The PRD content is ingested from user input or by reading files in the local workspace.
- Boundary markers: The instructions do not specify any delimiters or safety prompts to isolate the PRD content from the agent's core instructions.
- Capability inventory: The skill allows the agent to read project files (codebase exploration) and write Markdown files to the
./plans/directory. - Sanitization: No sanitization is performed on the PRD content or the resulting feature names, which could potentially be exploited for path manipulation or to influence the agent's logic during codebase exploration.
Audit Metadata