product-manager
Pass
Audited by Gen Agent Trust Hub on Jul 3, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface because it is designed to ingest and process untrusted external documents provided by users.
- Ingestion points: External document content is ingested in '模式一：文档评审' (Mode 1: Document Review) and '模式二：PRD创作' (Mode 2: PRD Creation) as defined in SKILL.md.
- Boundary markers: No specific delimiters or instructions to ignore instructions within user-provided documents are present.
- Capability inventory: The skill uses the web search tool for competitive analysis and generates .docx files for PRD delivery.
- Sanitization: No explicit sanitization or filtering of document content is implemented.
- [SAFE]: The skill instructions provide a highly structured and professional framework for product management tasks using industry-standard models like RICE and KANO.
- [SAFE]: The skill defines explicit boundaries for its use cases, referring users to other specialized agents for technical implementation or deep research.
- [SAFE]: No obfuscation, hidden commands, hardcoded credentials, or unauthorized network requests were found in the skill metadata or supporting documentation.
Audit Metadata