Actions7
- AI Advice From Graph Actions
- Knowledge Graph Actions
Overview
The "InfraNodus Graph RAG" node integrates with the InfraNodus API to work with knowledge graphs and obtain AI-generated advice based on those graphs. Specifically, the AI Advice From Graph resource allows querying existing InfraNodus graph experts to generate insights, prompts, questions, or summaries using AI models.
This node is beneficial in scenarios where users want to:
- Extract meaningful advice or answers from complex knowledge graphs.
- Generate AI-driven prompts or questions related to a specific graph.
- Summarize graph content highlighting gaps, keywords, and discourse optimization.
- Leverage different AI models for tailored reasoning and response quality.
Practical examples include:
- Asking an expert system built on a knowledge graph for recommendations or clarifications.
- Generating discussion topics or research questions from a graph of interconnected ideas.
- Summarizing large datasets represented as graphs to identify missing information or key themes.
Properties
| Name | Meaning |
|---|---|
| Graph Name | The name of the existing InfraNodus graph expert to use (required). |
| Prompt | The prompt text to send to the AI model when using the prompt operation. |
| Request Mode | The type of AI request (hidden, default: "response"). |
| AI Topics | Whether to generate AI topic names for graph clusters (hidden, default: true). |
| Discourse Optimization | Use parts of the graph that optimize the discourse; options relate to how the graph data is processed. |
| Do Not Save Graph | Whether to avoid saving a new graph after the operation (hidden, default: true). |
| Add Graph Stats | Whether to add graph statistics to the AI response (hidden, default: true). |
| Include Graph | Whether to include the graph data in the response (hidden, default: false). |
| Include Extended Graph Summary | Whether to include detailed info about gaps and keywords in the summary (options: Yes/No, default: Yes). |
| Model to Use | Select which AI model to use for generating the response. Options include Claude Opus, Gemini, GPT-4o, GPT-5 variants, etc. |
Output
The node outputs JSON data containing the AI-generated advice or summary based on the queried graph. The structure typically includes:
- The AI response text (e.g., answers, prompts, summaries).
- Optional graph statistics if enabled.
- Optionally, the graph data itself if requested.
- Extended summary details about content gaps and keywords when enabled.
If binary data were involved (not indicated here), it would represent files or media related to the graph or AI output, but this node focuses on JSON textual responses.
Dependencies
- Requires an active InfraNodus API key credential configured in n8n.
- Connects to the InfraNodus API endpoint at
https://infranodus.com/api/v1. - Uses selected AI models supported by InfraNodus for generating advice.
- No other external dependencies are indicated.
Troubleshooting
- Missing or invalid API credentials: Ensure the InfraNodus API key is correctly set up in n8n credentials.
- Graph not found error: Verify the "Graph Name" matches an existing InfraNodus graph expert.
- Invalid model selection: Confirm the chosen AI model is supported and spelled correctly.
- Empty or unexpected responses: Check if the graph contains sufficient data and that the prompt or query is well-formed.
- Network or API errors: Validate network connectivity and InfraNodus service status.
Links and References
- InfraNodus Official Website
- InfraNodus API documentation (refer to InfraNodus platform for API details)
- AI Models referenced (Claude, Gemini, GPT series) — consult respective providers for model capabilities and usage guidelines