prd-to-plan
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external, user-provided data in the form of Product Requirement Documents (PRDs) to generate implementation plans.
- Ingestion points: The skill ingests PRD content directly from the user's prompt or from specified files.
- Boundary markers: While the instructions state that external data should be treated as "hostile," there are no explicit technical delimiters (such as XML tags or specific guardrail sequences) defined to separate the untrusted PRD content from the agent's instructions.
- Capability inventory: The agent is instructed to explore the codebase (file read access) and write plans to the local filesystem (./plans/ folder). It also mentions executing terminal actions for testing.
- Sanitization: The instructions mention redacting secrets or PII before sharing context with subagents, but no specific sanitization or validation of the PRD content itself is required before it is processed by the LLM.
Audit Metadata