The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill performs network requests to down.mptext.top to retrieve article lists and full HTML content. This behavior is consistent with the skill's primary stated purpose of article aggregation.
[INDIRECT_PROMPT_INJECTION]: The skill ingests external content from the web, which creates a potential surface for indirect prompt injection.
Ingestion points: Article HTML is downloaded from the down.mptext.top API in scripts/fetch_articles.py.
Boundary markers: No specific boundary markers or delimiters are added to the parsed content.
Capability inventory: The skill has the ability to read from the network and write files to the local filesystem.
Sanitization: The script includes a ContentExtractor class that filters out script, style, and noscript tags, and optionally uses html2text for Markdown conversion, which significantly reduces the risk of active content injection.
[COMMAND_EXECUTION]: The tool uses the pathlib library for filesystem operations and includes a safe_filename function that sanitizes article titles by removing special characters and limiting length, mitigating risks such as directory traversal.

wechat-article-aggregator