wechat-article-writer
Pass
Audited by Gen Agent Trust Hub on Aug 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The script
wechat-full.jsutilizeschild_process.execSyncto interact with the system clipboard (pbcopy,xclip,clip) and to open the local HTML results in the default browser (open,xdg-open,start). These operations are used for local productivity enhancement. - [EXTERNAL_DOWNLOADS]: The skill instructs the user to install the Playwright library and the Chromium browser engine, which are downloaded from official package registries and browser distributions to enable automated screenshot capabilities.
- [PROMPT_INJECTION]: The skill processes user-supplied Markdown content to generate article structures. It currently lacks boundary markers or sanitization logic specifically designed to mitigate indirect prompt injection if the output is consumed by an LLM.
- Ingestion points: Markdown files are read from the local file system as specified by the user via command-line arguments in
wechat-full.js. - Boundary markers: None identified in the parsing or rendering logic.
- Capability inventory: The skill has capabilities for local file system reading and writing, shell command execution for system utilities, and headless browser control.
- Sanitization: The code performs structural parsing into JSON objects before rendering to HTML, but does not sanitize the text content for potential embedded instructions.
Audit Metadata