OneAI icon

OneAI

Execute OneAI related operations

Actions2

Overview

This node integrates with the OneAI platform to perform Retrieval-Augmented Generation (RAG) queries. It allows users to search for relevant information by providing either a natural language query or keywords, optionally filtered by workspace and content type. The node supports advanced search configurations such as similarity thresholds, metadata filtering, and different RAG modes (hybrid, embedding-only, or full-text only). This is useful in scenarios where you want to retrieve contextually relevant documents or Q&A pairs from a knowledge base to augment AI-generated responses or to power intelligent search features.

Practical examples:

  • Asking "What is artificial intelligence?" to retrieve top relevant documents and Q&A entries from multiple workspaces.
  • Searching with keywords like "AI|Machine Learning|Deep Learning" to find related content across selected workspaces.
  • Using hybrid RAG mode with custom weights to balance embedding and full-text search results for more precise retrieval.

Properties

Name Meaning
Query The question or content to search for. Cannot be empty if Keywords are also empty. Example: "What is artificial intelligence?"
Keywords Keywords separated by "
Workspaces (Comma-Separated) Comma-separated list of workspace names or IDs to restrict the search scope. Leave empty to search all accessible workspaces. Example: "workspace_alpha,workspace_beta_id"
RAG Object Type Specify the content type to search: All (Q&A and Documents), Q&A Only, or Documents Only
Top K Results Number of search results to return. Minimum value is 1. Default is 3
Minimum Similarity Minimum similarity score for results, ranging from 0 to 1. Default is 0.8
Metadata Filter Metadata filter type to use. Currently only supports "default". Leave empty if not used
Metadata Search Type Specify the metadata search type: Filtration or Sort
RAG Mode Specify the RAG search mode: Hybrid (Embedding and Full Text), Embedding Only, or Full Text Only
Weights (JSON for Hybrid Mode) JSON object specifying RRF weights for different search indices (e.g., {"Embedding": 0.9, "FullText": 0.8}). Effective only when RAG mode is Hybrid and this field is not empty
Reranker Specify the reranker model to use. Leave empty if not used

Output

The node outputs JSON data containing the search results returned by the OneAI RAG API. The structure typically includes an array of matched documents or Q&A entries, each with associated metadata such as similarity scores, content snippets, and identifiers. This output can be used downstream in workflows for further processing, display, or decision-making.

No binary data output is indicated for this operation.

Dependencies

  • Requires an API key credential for authenticating with the OneAI platform.
  • The node sends HTTP POST requests to the OneAI API endpoint /v1/openapi/rag.
  • The base URL for the API is configurable via credentials.
  • Proper network access to the OneAI service is necessary.

Troubleshooting

  • Empty Query and Keywords: The node requires at least one of Query or Keywords to be non-empty. If both are empty, the API call will likely fail or return no results.
  • Invalid JSON in Weights: When using the Weights property in Hybrid mode, ensure the JSON string is valid. Malformed JSON will cause parsing errors.
  • API Authentication Errors: Ensure the API key credential is correctly configured and has sufficient permissions.
  • Workspace Filtering Issues: If specifying workspaces, verify that the workspace names or IDs are correct and accessible by the API key.
  • Similarity Threshold Too High: Setting Minimum Similarity too close to 1 may result in no results being returned.
  • Unsupported Metadata Filter: Currently, only "default" is supported for metadata filtering; other values may cause errors.

Links and References

Discussion