md2wechat
Pass
Audited by Gen Agent Trust Hub on Mar 31, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill relies on the 'md2wechat' CLI tool to execute various tasks, including article conversion, preview generation, and configuration management. It encourages the use of discovery commands to verify available resources before processing content.
- [EXTERNAL_DOWNLOADS]: The skill communicates with external APIs, specifically the official WeChat Official Account API for uploading materials and managing drafts. It also supports fetching images from remote URLs and interacting with user-configured image generation services.
- [CREDENTIALS_UNSAFE]: The skill requires WeChat API credentials ('WECHAT_APPID', 'WECHAT_SECRET') for publishing features. It appropriately instructs the agent and user to manage these secrets through environment variables or local configuration files in the user's home directory, avoiding insecure hardcoding.
- [DATA_EXFILTRATION]: The skill's primary function involves uploading local Markdown content, titles, and images to WeChat's infrastructure. This transmission is consistent with the skill's stated purpose and is initiated only upon explicit user request.
- [PROMPT_INJECTION]: As the skill ingests and processes Markdown files and user-defined metadata (titles, digests) for conversion into HTML, it possesses a surface for indirect prompt injection. However, the provided guidelines for inspection and previewing mitigate the risk of unintended output.
Audit Metadata