The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local Python script scripts/fetch_gzh_trends.py to process search queries and format results. This is the primary functional component of the skill.
[EXTERNAL_DOWNLOADS]: The Python script makes HTTP POST requests to https://redfox.hk/story/api/gzh/search/hotArticleNew to retrieve article data. This domain belongs to the skill's vendor and is necessary for its stated purpose.
[CREDENTIALS_UNSAFE]: The skill uses the REDFOX_API_KEY environment variable for authentication. This is a standard and secure method for handling API credentials in agentic workflows, avoiding hardcoded secrets.
[PROMPT_INJECTION]: The SKILL.md file contains detailed instructions to guide the AI's behavior, including strict formatting rules and prohibitions against unauthorized analysis. These serve as behavioral safeguards rather than malicious overrides.

wechat-search