ragflow_retrieval_by_name
Retrieve relevant document chunks from specified datasets using semantic search and similarity scoring to answer queries.
Instructions
Retrieve document chunks by dataset names using the retrieval API. Returns raw chunks with similarity scores.
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| dataset_names | Yes | List of names of the datasets/knowledge bases to search (e.g., ['BASF', 'Legal']) | |
| query | Yes | Search query or question | |
| document_name | No | Optional document name to filter results to specific document | |
| top_k | No | Number of chunks for vector cosine computation. Defaults to 1024. | |
| similarity_threshold | No | Minimum similarity score for chunks (0.0 to 1.0). Defaults to 0.2. | |
| page | No | Page number for pagination. Defaults to 1. | |
| page_size | No | Number of chunks per page. Defaults to 10. | |
| use_rerank | No | Whether to enable reranking for better result quality. Default: false (uses vector similarity only). | |
| deepening_level | No | Level of DSPy query refinement (0-3). 0=none, 1=basic refinement, 2=gap analysis, 3=full optimization. Default: 0 |