task-prd-creator
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill performs read-only operations on the local file system using
catandgrepto analyze existing project architecture, rules, and implementation patterns. - [EXTERNAL_DOWNLOADS]: Fetches library documentation from external sources via the
Context7 MCPtool. The workflow references fetching content from official repositories such asvercel/next.jsandnestjs/docs.nestjs.comto ensure planning is based on current documentation. - [INDIRECT_PROMPT_INJECTION]: The skill ingests user input and existing project content to generate new documentation files, creating a potential surface for indirect injection.
- Ingestion points: User feature requests and existing project files in
.agent/SYSTEM/and.agent/EXAMPLES/. - Boundary markers: Not explicitly defined within the generated Task or PRD templates, though the workflow includes a mandatory human-in-the-loop approval step.
- Capability inventory: Local file reading, directory searching, and documentation retrieval through MCP tools.
- Sanitization: There are no documented steps for sanitizing or validating the content retrieved from project files or user input before processing.
Audit Metadata