zj-to-prd
Pass
Audited by Gen Agent Trust Hub on Jul 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill follows a structured workflow to analyze the codebase and conversation context to produce documentation without any malicious intent or hidden commands.
- [SAFE]: No obfuscation, unauthorized data access, or remote code execution patterns were detected in the instructions or template.
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests untrusted conversation context to generate output that is subsequently published to an external system (issue tracker).
- Ingestion points: Conversation context and repository data.
- Boundary markers: No specific delimiters are used to isolate the conversation context from the PRD generation instructions.
- Capability inventory: The skill is capable of publishing content to the project's issue tracker (writing data externally).
- Sanitization: No explicit sanitization of the synthesized content is performed before the publishing step.
Audit Metadata