extract_graph_v1
Extracts entities, relationships, and conclusions from document chunks to build structured knowledge graphs for academic literature analysis.
Instructions
抽取结构化图谱要素 (Async Parallel)
从文档的 chunks 中抽取实体、关系和结论，写入 GraphRAG 表。 使用并行处理以加快速度。
Args: concurrency: 并发请求数，默认 60。OpenRouter 支持较高并发 (500 RPS)。
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| doc_id | No | ||
| chunk_ids | No | ||
| mode | No | high_value_only | |
| max_chunks | No | ||
| llm_model | No | ||
| min_confidence | No | ||
| dry_run | No | ||
| concurrency | No |