Actions14
- Agent Actions
- File Actions
- Image Actions
- Prompt Actions
- RAG Actions
Overview
The node integrates with the Straico API to perform various operations related to Retrieval-Augmented Generation (RAG) resources. Specifically, for the RAG Prompt Completion operation, it sends a user-provided prompt to a specified RAG resource and retrieves a generated completion based on that prompt. This is useful in scenarios where you want to leverage a knowledge base or document collection indexed by the RAG system to generate context-aware responses or completions.
Practical examples include:
- Querying a custom knowledge base to get detailed answers.
- Generating text completions informed by specific documents uploaded to the RAG.
- Enhancing chatbots or virtual assistants with domain-specific information retrieval and generation.
Properties
| Name | Meaning |
|---|---|
| RAG ID | The unique identifier of the RAG resource to which the prompt will be sent. |
| Prompt | The input text prompt for which the completion is requested. |
| Model | The model identifier used to generate the completion. |
| Search Type | The method used to search within the RAG documents before generating the completion. Options: Similarity, Mmr, Similarity_score_threshold |
| K | Number of top relevant chunks/documents to consider for the completion. |
| Fetch_k | Number of chunks/documents to fetch initially before filtering down to K. |
| Lambda_mult | A weighting multiplier parameter used in the search algorithm (e.g., for MMR). |
| Score_threshold | Threshold score to filter out less relevant chunks/documents during search. |
Output
The output JSON contains the response from the Straico API for the prompt completion request. It typically includes fields such as:
id: Identifier of the completion request.message: The generated completion text.model: The model used for generation.usage: Information about token usage or cost.created_at: Timestamp of the completion creation.
If the "Simplify" option is enabled, the output is a simplified object containing only these key fields instead of the full raw response.
No binary data output is produced by this operation.
Dependencies
- Requires an API key credential for authenticating requests to the Straico API.
- The node makes HTTP POST requests to the endpoint:
https://api.straico.com/v0/rag/{ragId}/prompt - Content-Type header for the request is
application/x-www-form-urlencoded. - The node depends on n8n's HTTP request helper with authentication support.
Troubleshooting
- Missing Binary Data Error: Not applicable for this operation since it does not handle binary inputs.
- Invalid RAG ID: If the provided RAG ID does not exist or is incorrect, the API will likely return an error. Verify the RAG ID is correct and accessible.
- Model Parameter Required: The model property is mandatory; omitting it will cause the request to fail.
- Search Type and Parameters: Using invalid values for search type or numeric parameters like
k,fetch_k,lambda_mult, orscore_thresholdmay result in unexpected behavior or errors. Use valid options and reasonable numeric ranges. - API Authentication Errors: Ensure the API key credential is correctly configured in n8n and has sufficient permissions.
- Simplify Option: If enabled, the output is reduced to essential fields. If you expect full details, disable this option.
Links and References
- Straico API Documentation (general): https://straico.com/docs (hypothetical link)
- Retrieval-Augmented Generation concepts: https://en.wikipedia.org/wiki/Retrieval-augmented_generation
- n8n HTTP Request Node documentation: https://docs.n8n.io/nodes/n8n-nodes-base.httpRequest/