wechat-article-writer

Pass

Audited by Gen Agent Trust Hub on Feb 27, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: A thorough audit of the skill instructions and reference materials found no evidence of malicious intent. The skill follows established patterns for content generation and research.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface due to its reliance on external web data fetched via web_fetch and exa:web_search_exa. 1. Ingestion points: web_fetch and exa:web_search_exa calls in SKILL.md used for material analysis and research. 2. Boundary markers: Not present; the skill does not explicitly instruct the model to ignore instructions found within external content. 3. Capability inventory: The skill utilizes search, web fetching, and memory MCP tools. 4. Sanitization: No explicit validation or filtering of fetched web content is performed.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 27, 2026, 01:11 PM