mj-adapt
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references font resources from Google Fonts (
fonts.googleapis.com) in its HTML templates (assets/xhs-ending-template.html). This is a standard and safe use of a well-known service for styling purposes.\n- [COMMAND_EXECUTION]: The skill instructions inSKILL.mddirect the agent to execute local JavaScript utility scripts (generate-cover.js,generate-images.js,generate-xhs-slides.js) using Node.js to perform rendering and image generation tasks. These scripts are included within the skill package.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied Markdown content to generate formatted output for various platforms.\n - Ingestion points: User-provided article content or file paths processed in
SKILL.mdStep 1.\n - Boundary markers: Explicit instructions are provided to the agent to preserve original content without modification or re-interpretation (e.g., "严禁修改原文内容").\n
- Capability inventory: File system writes to the
output/directory and execution of local Node.js scripts for image rendering.\n - Sanitization: The skill converts text into static HTML and PNG images, which significantly mitigates the potential for injected instructions to influence downstream systems.
Audit Metadata