The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data by fetching WeChat articles from the internet. The resulting Markdown is provided to the agent, which could potentially contain instructions aimed at overriding the agent's behavior.
Ingestion points: The script scripts/fetch_weixin.py fetches data from external URLs provided in the prompt.
Boundary markers: The skill does not use specific delimiters or instructions to treat the fetched content as untrusted data.
Capability inventory: The skill allows writing files to the filesystem and executing Python scripts.
Sanitization: Content is filtered by HTML selector and converted to Markdown, which removes script tags but preserves text.
[EXTERNAL_DOWNLOADS]: The skill setup process involves downloading the Chromium browser binary using the playwright install chromium command. This is a standard operation for the Playwright library.
[COMMAND_EXECUTION]: The skill logic relies on the agent executing the scripts/fetch_weixin.py script via the shell to perform its primary function. The script includes basic filename sanitization for Windows environments.

weixin-fetch