The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The file send_wecom_summary.py contains a hardcoded WeCom (Enterprise WeChat) webhook URL including a functional authentication key (key=45548262-1f73-40a9-a33f-eba95e934082). Hardcoding secrets exposes the bot's communication channel to unauthorized use.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from GitHub project READMEs and incorporates it into summaries processed by the agent, creating a vector for indirect instruction injection.
Ingestion points: fetch_trending.py retrieves raw README text from arbitrary repositories found on the GitHub Trending page.
Boundary markers: No delimiters or safety instructions are used to separate external content from the agent's core instructions.
Capability inventory: The skill uses urllib.request for network access and writes to the local filesystem via json.dump.
Sanitization: While send_wecom_summary.py includes basic markdown character stripping, it does not validate or filter the content for potentially malicious instructions.
[COMMAND_EXECUTION]: The skill requires the agent to execute multiple Python scripts (fetch_trending.py and send_wecom_summary.py) to process data and perform network operations.

github-trending