wechat-article
Pass
Audited by Gen Agent Trust Hub on Jul 4, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructions involve retrieving external content to support the research phase.
- Evidence: It directs agents to download original papers (PDFs) and perform web searches for technical, industry, and competitive analysis materials.
- [COMMAND_EXECUTION]: The workflow requires the use of command-line utilities to process visual assets for the articles.
- Evidence: It specifies the use of
pdftoppmvia shell commands to convert PDF pages into high-resolution images. - [PROMPT_INJECTION]: As a research-oriented tool, the skill has an inherent surface for indirect prompt injection via external data ingestion.
- Ingestion points: Content from technical PDFs and web search results are processed by the agent in SKILL.md.
- Boundary markers: No specific delimiters or "ignore instructions" directives are provided for handling external content.
- Capability inventory: The agent has capabilities for file writing, shell command execution, and multi-agent delegation.
- Sanitization: The instructions do not describe specific sanitization or validation logic for external data inputs.
Audit Metadata