hz_run_pipeline
Execute a content pipeline to fetch, score, filter, enrich, and summarize data from multiple sources for analysis.
Instructions
一键执行抓取->打分->过滤->富化->摘要。
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| hours | No | ||
| languages | No | ||
| threshold | No | ||
| horizon_path | No | ||
| config_path | No | ||
| sources | No | ||
| enrich | No | ||
| topic_dedup | No | ||
| save_to_horizon_data | No |
Implementation Reference
- horizon_mcp/server.py:262-288 (handler)The `hz_run_pipeline` function is defined as an MCP tool handler in `horizon_mcp/server.py`, which delegates the execution to `service.run_pipeline`.
async def hz_run_pipeline( hours: int = 24, languages: list[str] | None = None, threshold: float | None = None, horizon_path: str | None = None, config_path: str | None = None, sources: list[str] | None = None, enrich: bool = True, topic_dedup: bool = True, save_to_horizon_data: bool = False, ) -> dict[str, Any]: """一键执行抓取->打分->过滤->富化->摘要。""" return await _run_tool( "hz_run_pipeline", lambda: service.run_pipeline( hours=hours, languages=languages, threshold=threshold, horizon_path=horizon_path, config_path=config_path, sources=sources, enrich=enrich, topic_dedup=topic_dedup, save_to_horizon_data=save_to_horizon_data, ), )