The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Python scripts (ann_query.py and top_n_pipeline.py) to automate data collection and processing. These scripts utilize the Python standard library to perform specific financial research tasks such as fetching stock quotes and sorting market gainers.
[EXTERNAL_DOWNLOADS]: The skill retrieves financial market data and corporate announcements from well-known Chinese services, including East Money (eastmoney.com), Sina (sina.com.cn), and CNINFO (cninfo.com.cn). These operations are consistent with the skill's stated purpose and target established, reputable data providers.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external financial reports and company announcements, which constitutes an ingestion surface for untrusted content.
Ingestion points: Corporate announcements and market data fetched via ann_query.py and top_n_pipeline.py are loaded into the agent's context.
Boundary markers: The skill implements a robust "adversarial validation" phase using "fresh-context" agents to independently verify findings, which serves as a significant mitigation against data-driven hallucinations or embedded instructions.
Capability inventory: The skill can execute shell commands (Python scripts), write local CSV files, perform network requests to financial domains, and orchestrate sub-agents.
Sanitization: Content is handled as structured data within CSV files, and the skill emphasizes accuracy and source-tracing in its "Highest Discipline" instructions.

daymade-sector-research