The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the AI agent to execute a Python script (scripts/scrape_wechat_mp.py) to perform its primary function. This script is part of the skill's distribution and uses the standard Python library.
[EXTERNAL_DOWNLOADS]: The Python script performs network operations to fetch HTML content from mp.weixin.qq.com. It also downloads assets (images, background images) found within the article content to a local directory (~/wechat-mp-scraper-runs). The script includes a validation check to restrict the article URL to the official WeChat domain.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection (Category 8) because it ingests and processes untrusted content from external WeChat articles.
Ingestion points: scripts/scrape_wechat_mp.py fetches article HTML and assets, which are then saved as content.md, content.json, and report.md.
Boundary markers: The instructions do not define clear delimiters or warnings to the agent to ignore instructions embedded within the scraped content.
Capability inventory: The agent has the capability to execute the Python script via python3 and read the generated output files.
Sanitization: The script performs text normalization and HTML unescaping but does not sanitize the content for potential malicious prompt instructions targeted at the agent.

wechat-mp-scraper