prd-to-ux
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEPROMPT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill contains 'The Iron Law' and 'Red Flags' sections which provide explicit instructions to disregard certain user inputs that might attempt to skip steps.
- Evidence: The text instructs the agent to ignore requests like 'I'm in a hurry', 'Just give me screens', or 'Skip the analysis'.
- Context: These are process-oriented guardrails intended to ensure output quality rather than malicious attempts to bypass agent safety protocols.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it ingests data from external documents and has file-writing capabilities.
- Ingestion points: The skill is designed to read and translate 'PRDs, feature specs, or product requirements' from external files.
- Boundary markers: There are no defined delimiters or instructions to ignore potential commands embedded within the source documents.
- Capability inventory: The skill specifies writing output to a file in the local directory (e.g.,
{prd-basename}-ux-spec.md). - Sanitization: No sanitization or validation of the input document content is described before it is processed and rewritten.
Audit Metadata