prd-to-ux
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates entirely as a documentation and prompt-generation assistant. It follows a strictly defined multi-step pipeline and uses pre-defined templates for all outputs.
- [SAFE]: File operations are limited to writing Markdown artifacts to a specific sub-directory (
docs/prd-to-ux/) within the user's workspace. The skill does not attempt to access sensitive system files or environment variables. - [SAFE]: The skill explicitly defines tool-agnostic behavior and does not invoke any external rendering tools, APIs, or network services. It functions as a text-to-text transformation tool.
- [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests user-supplied PRDs and product ideas. However, this risk is mitigated by the skill's limited capabilities, which are restricted to writing descriptive text files without execution or tool invocation. This surface is inherent to the skill's primary purpose and lacks a viable exploitation path.
- Ingestion points: User-provided PRD file paths or pasted text content processed in SKILL.md and Step 1.
- Boundary markers: The skill uses specific Markdown templates to structure output, providing implicit boundaries between user content and agent instructions.
- Capability inventory: File-write operations to the local filesystem for artifact generation.
- Sanitization: Input is validated for completeness (PRD Validation Gate) but not sanitized for injection patterns.
Audit Metadata