> ## Documentation Index > Fetch the complete documentation index at: https://docs.threetone.in/llms.txt > Use this file to discover all available pages before exploring further. # Query agent knowledge base > Run the agent's current RAG retrieval for an ad hoc query and return ranked chunks without creating a conversation ## OpenAPI ````yaml /api-reference/openapi.json post /v1/convai/agents/{agent_id}/knowledge-base/rag-query openapi: 3.1.0 info: title: Threetone API description: >- APIs for the Threetone Agents Platform and shared workspace resources. Authenticate programmatic requests with an API key in the x-api-key header. version: '1.0' servers: - url: https://api.threetone.in description: Production security: [] paths: /v1/convai/agents/{agent_id}/knowledge-base/rag-query: post: tags: - Agents Platform summary: Query agent knowledge base description: >- Run the agent's current RAG retrieval for an ad hoc query and return ranked chunks without creating a conversation operationId: query_agent_knowledge_base_rag_route parameters: - name: agent_id in: path required: true schema: type: string description: The id of an agent. This is returned on agent creation. examples: - agent_3701k3ttaq12ewp8b7qv5rfyszkz embed: true title: Agent Id description: The id of an agent. This is returned on agent creation. - name: branch_id in: query required: false schema: anyOf: - type: string - type: 'null' description: The ID of the branch to use examples: - agtbranch_0901k4aafjxxfxt93gd841r7tv5t title: Branch Id description: The ID of the branch to use - name: x-api-key in: header required: false schema: anyOf: - type: string - type: 'null' description: >- Your API key for programmatic access. Send it in the x-api-key header; you can create and manage keys from the Developers page. title: X-Api-Key description: >- Your API key for programmatic access. Send it in the x-api-key header; you can create and manage keys from the Developers page. requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/AgentKnowledgeBaseRagQueryRequestModel' responses: '200': description: Successful Response content: application/json: schema: $ref: '#/components/schemas/AgentKnowledgeBaseRagQueryResponseModel' '422': description: Validation Error content: application/json: schema: $ref: '#/components/schemas/HTTPValidationError' components: schemas: AgentKnowledgeBaseRagQueryRequestModel: properties: query: type: string minLength: 1 title: Query description: Query to run against the agent's knowledge base RAG index. examples: - How do I reset my password? use_agent_defaults: type: boolean title: Use Agent Defaults description: >- When true (the default), retrieval uses the agent's own RAG settings, reproducing exactly what the agent would retrieve. Set to false to retrieve with neutral default RAG settings instead (the agent's embedding model is always kept, since it determines which vector index exists). Useful for auditing the knowledge base independently of how a particular agent is tuned. default: true max_documents_length: anyOf: - type: integer maximum: 50000 exclusiveMinimum: 0 - type: 'null' title: Max Documents Length description: >- Optional maximum total character length of document chunks returned. Overrides the selected RAG settings for this query only. max_retrieved_rag_chunks_count: anyOf: - type: integer maximum: 20 exclusiveMinimum: 0 - type: 'null' title: Max Retrieved Rag Chunks Count description: >- Optional maximum number of document chunks retrieved. Overrides the selected RAG settings for this query only. type: object required: - query title: AgentKnowledgeBaseRagQueryRequestModel AgentKnowledgeBaseRagQueryResponseModel: properties: retrieval_query: type: string title: Retrieval Query description: Raw query used for retrieval (echoes the request query). chunks: items: $ref: '#/components/schemas/AgentKnowledgeBaseRagChunkResponseModel' type: array title: Chunks description: >- Ranked chunks the agent would retrieve, after distance and length filtering. type: object required: - retrieval_query - chunks title: AgentKnowledgeBaseRagQueryResponseModel example: chunks: - chunk_id: chunk_498cjVigY5qzO86Huf0OWal content_format: html document_id: doc_J3Pbu5gP6NNKBscdCdwB document_name: Account FAQ document_type: url source_url: >- https://example.com/account-faq?utm_medium=agents_citation&utm_source=elevenlabs text: To reset your password, open Settings and choose Security... vector_distance: 0.21 retrieval_query: How do I reset my password? HTTPValidationError: properties: detail: items: $ref: '#/components/schemas/ValidationError' type: array title: Detail type: object title: HTTPValidationError AgentKnowledgeBaseRagChunkResponseModel: properties: document_id: type: string title: Document Id description: ID of the source knowledge base document. document_name: type: string title: Document Name description: Name of the source knowledge base document. source_url: anyOf: - type: string - type: 'null' title: Source Url description: >- Tracked source URL for URL documents, or null for other document types. chunk_id: type: string title: Chunk Id description: ID of the retrieved chunk. text: type: string title: Text description: Text content of the retrieved chunk. vector_distance: anyOf: - type: number - type: 'null' title: Vector Distance description: Similarity distance when exposed by the retrieval strategy. content_format: $ref: '#/components/schemas/ContentFormat' description: >- Format of the chunk text. Markdown chunks contain raw markdown; HTML chunks contain HTML. document_type: $ref: '#/components/schemas/KnowledgeBaseDocumentType' description: Type of the source knowledge base document. type: object required: - document_id - document_name - source_url - chunk_id - text - vector_distance - content_format - document_type title: AgentKnowledgeBaseRagChunkResponseModel ValidationError: properties: loc: items: anyOf: - type: string - type: integer type: array title: Location msg: type: string title: Message type: type: string title: Error Type type: object required: - loc - msg - type title: ValidationError ContentFormat: type: string enum: - html - markdown title: ContentFormat description: >- Canonical representation of a knowledge base document's stored content. HTML is the legacy default; documents created before this field existed are interpreted as HTML. default: html KnowledgeBaseDocumentType: type: string enum: - file - url - text - folder title: KnowledgeBaseDocumentType ````