The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user or agent to install the miku_ai Python package using pip install miku_ai --break-system-packages. This is an external dependency from an unverified third-party source.
[COMMAND_EXECUTION]: The skill uses python3 -c to execute an inline asynchronous Python script that imports the miku_ai library and performs the article search. This involves running code that relies on external dependencies and user-provided keywords.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection (Category 8).
Ingestion points: Data is ingested from the get_wexin_article function, which fetches titles and digests from external WeChat articles.
Boundary markers: There are no boundary markers or instructions to ignore embedded commands within the fetched article content.
Capability inventory: The skill uses asyncio.run and python3 subprocess execution to run the search logic.
Sanitization: The fetched article metadata (title, digest) is printed directly to the output without sanitization or escaping, which could lead to an agent following instructions hidden in article summaries.

wechat-search