RAG-powered Conversational Agent

POST

Engage with an intelligent RAG-powered conversational agent for complex information retrieval and analysis.

This advanced endpoint combines retrieval-augmented generation (RAG) with a conversational AI agent to provide detailed, context-aware responses based on your document collection. The agent can:

  • Maintain conversation context across multiple interactions
  • Dynamically search and retrieve relevant information from both vector and knowledge graph sources
  • Break down complex queries into sub-questions for comprehensive answers
  • Cite sources and provide evidence-based responses
  • Handle follow-up questions and clarifications
  • Navigate complex topics with multi-step reasoning

Key Features:

  • Hybrid search combining vector and knowledge graph approaches
  • Contextual conversation management with conversation_id tracking
  • Customizable generation parameters for response style and length
  • Source document citation with optional title inclusion
  • Streaming support for real-time responses
  • Branch management for exploring different conversation paths

Common Use Cases:

  • Research assistance and literature review
  • Document analysis and summarization
  • Technical support and troubleshooting
  • Educational Q&A and tutoring
  • Knowledge base exploration

The agent uses both vector search and knowledge graph capabilities to find and synthesize information, providing detailed, factual responses with proper attribution to source documents.

Request

This endpoint expects an object.
messageobjectOptional

Current message to process

search_modeenumOptional
Allowed values: basicadvancedcustom

Default value of custom allows full control over search settings.

Pre-configured search modes: basic: A simple semantic-based search. advanced: A more powerful hybrid search combining semantic and full-text. custom: Full control via search_settings.

If filters or limit are provided alongside basic or advanced, they will override the default settings for that mode.

search_settingsobjectOptional

The search configuration object. If search_mode is custom, these settings are used as-is. For basic or advanced, these settings will override the default mode configuration.

Common overrides include filters to narrow results and limit to control how many results are returned.

rag_generation_configobjectOptional

Configuration for RAG generation

task_prompt_overridestringOptional

Optional custom prompt to override default

include_title_if_availablebooleanOptional

Include document titles in responses when available

conversation_idstringOptional

ID of the conversation

branch_idstringOptional

ID of the conversation branch

messageslist of objectsOptionalDeprecated

List of messages (deprecated, use message instead)

Response

Successful Response

resultsobject

Errors