The Agent Skills Directory

[PROMPT_INJECTION]: The skill provides access to several data sources that ingest untrusted, user-generated content, creating a surface for indirect prompt injection.
Ingestion points: Data is retrieved from financial news sites and public Q&A platforms (e.g., irm_qa_sh, irm_qa_sz, news, major_news) as documented in the references/ directory.
Boundary markers: Absent. The skill instructions do not specify delimiters or provide guidance for the agent to ignore instructions embedded within the retrieved data.
Capability inventory: The skill environment allows for Python code execution via provided scripts (scripts/stock_data_example.py, scripts/fund_data_example.py) and standard library usage.
Sanitization: Absent. There is no evidence of filtering or escaping external content before it enters the agent's context.
[EXTERNAL_DOWNLOADS]: The skill instructs the agent or user to install the tushare Python package from the Tsinghua University PyPI mirror, which is a standard and well-known repository.

tushare