> ## Documentation Index > Fetch the complete documentation index at: https://developers.datagrid.com/llms.txt > Use this file to discover all available pages before exploring further. # Converse > Converse with an AI Agent ## OpenAPI ````yaml post /converse openapi: 3.0.3 info: version: 0.1.1 title: Datagrid API description: Datagrid API servers: - url: https://api.datagrid.com/v1 security: - BearerAuth: [] paths: /converse: post: tags: - Converse summary: Converse description: Converse with an AI Agent operationId: Converse requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/ConverseRequestBody' responses: '201': description: Converse response content: application/json: schema: $ref: '#/components/schemas/ConverseResponse' '402': description: >- Credit limit exceeded. The organization does not have enough credits to perform this operation. Check your billing status and add credits to your account. '404': description: >- Not found. A referenced resource (agent, conversation, knowledge, or page) does not exist, or the authenticated user does not have permission to access it. '413': description: >- Payload too large. The input prompt exceeds the maximum allowed size. Reduce the length of your prompt and try again. '429': description: >- Rate limit exceeded. The request has been throttled because the rate limit for this endpoint has been reached. Check the `Retry-After` response header and retry after the specified number of seconds. content: application/json: schema: $ref: '#/components/schemas/RateLimitError' x-codeSamples: - lang: JavaScript source: >- import Datagrid from 'datagrid-ai'; const client = new Datagrid({ apiKey: process.env['DATAGRID_API_KEY'], // This is the default and can be omitted }); const converseResponse = await client.converse({ prompt: 'string' }); console.log(converseResponse); - lang: Python source: |- import os from datagrid_ai import Datagrid client = Datagrid( api_key=os.environ.get("DATAGRID_API_KEY"), # This is the default and can be omitted ) converse_response = client.converse( prompt="string", ) print(converse_response) components: schemas: ConverseRequestBody: type: object required: - prompt properties: prompt: oneOf: - type: string title: Text prompt description: A text prompt to send to the agent. - type: array title: Input item list description: | A list of one or many input items to the agent, containing different content types. items: $ref: '#/components/schemas/InputItem' config: description: >- Override the agent config for this converse call. This is applied as a partial override. nullable: true allOf: - $ref: '#/components/schemas/InputAgentConfig' - type: object properties: agent_tools: $ref: '#/components/schemas/AgentTools' description: Deprecated, use tools instead deprecated: true disabled_agent_tools: $ref: '#/components/schemas/AgentTools' description: >- Deprecated, use disabled_tools instead. If not provided - no tools are disabled. deprecated: true nullable: true mcp_servers: type: array description: > **BETA**: This feature is in beta and the schema may change. Array of MCP (Model Context Protocol) server configurations to enable for this converse call. Each MCP server provides additional tools that the agent can use during the conversation. Datagrid handles the full MCP lifecycle automatically: `initialize`, `notifications/initialized`, then `tools/list` / `tools/call` with `MCP-Session-Id` and `MCP-Protocol-Version` headers. nullable: true items: $ref: '#/components/schemas/InputMcpServer' agent_id: type: string description: >- The ID of the agent that should be used for the converse. When omitted and a `conversation_id` is provided, the conversation's existing agent assignments are preserved. When omitted without a `conversation_id`, a new conversation is created with the default agent. nullable: true agent_routing: description: >- Controls how the API selects which agent to use when routing is needed. This field is mutually exclusive with `agent_id`. When both `agent_id` and `agent_routing` are omitted and a `conversation_id` is provided, the conversation's existing agent assignments are preserved. nullable: true allOf: - $ref: '#/components/schemas/AgentRouting' conversation_id: type: string description: >- The ID of the present conversation to use. If it's not provided - a new conversation will be created. nullable: true stream: type: boolean description: >- Determines the response type of the converse. Response is the Server-Sent Events if stream is set to true. default: false nullable: true generate_citations: type: boolean description: >- Determines whether the response should include citations. When enabled, the agent will generate citations for factual statements. default: false nullable: true generate_title: type: boolean description: >- Determines whether generated_title metadata should be included. Defaults to false. generated_title is emitted only when this flag is explicitly true. default: false nullable: true include_steps: type: boolean description: >- When set to false, tool call and reasoning step events are omitted from SSE streams. Non-streaming responses always include the tool_calls and reasoning fields (as null when empty). default: true nullable: true chat_mode: type: string description: >- Controls how the agent processes the request for this turn. Matches the chat mode selector in the Datagrid web app: **Execute** (`full_agent`), **Extended** (`light_agent`), **Ask** (`llm_router`). When set to `auto`, the router jointly predicts the best agent and concrete mode (`full_agent` / `light_agent` / `llm_router`) per message. When set to a concrete mode, that mode is used directly. When omitted, the mode is determined by the `agent_model` in `config`. nullable: true enum: - auto - full_agent - light_agent - llm_router secret_ids: type: array description: >- Array of secret ID's to be included in the context. The secret value will be appended to the prompt but not stored in conversation history. items: type: string nullable: true text: type: object description: > Contains the format property used to specify the structured output schema (`text.format`). Structured output is supported for all `agent_model` values and `chat_mode` settings when `text.format` is provided (same JSON Schema mechanism everywhere). **Ask** in the web app maps to `chat_mode` `magpie-2.5-flash`;. nullable: true properties: format: type: object description: > The converse response will be a JSON string object, that adheres to the provided JSON schema. ```javascript const exampleJsonSchema = { $id: "movie_info", title: "movie_info", type: "object", properties: { name: { type: "string", description: "The name of the movie", }, director: { type: "string", description: "The director of the movie", }, release_year: { type: "number", description: "The year the movie was released", }, }, required: ["name", "director", "release_year"], additionalProperties: false, }; const response = await datagrid.converse({ prompt: "What movie won best picture at the 2001 oscars?", text: { format: exampleJsonSchema }, }); // Example response: "{ "name": "Gladiator", "director": "Ridley Scott", "release_year": 2000 }" const parsedResponse = JSON.parse(response.content[0].text); ``` user: description: >- Override user information for this converse call. This allows you to override the first_name, last_name, and email that will be used in the conversation context. nullable: true allOf: - $ref: '#/components/schemas/UserOverride' current_view_content: type: string description: >- A datagrid file URI pointing to content the user is currently viewing on screen (e.g., a web page, document, or dashboard rendered as markdown). The agent uses this context to resolve ambiguous queries like 'what is this about?' or 'review this'. The content is automatically summarized and made available to the agent. nullable: true reference_date: type: string format: date pattern: ^[0-9]{4}-[0-9]{2}-[0-9]{2}$ description: >- Optional deterministic reference date override in YYYY-MM-DD format. Must be a real calendar date (for example, rejects impossible dates like 2026-02-31). When set, the agent treats this date as today for relative date resolution and date context rendering. nullable: true ConverseResponse: allOf: - $ref: '#/components/schemas/ConversationMessage' - type: object properties: chat_mode: type: string description: >- The chat mode used for this response (web app: Execute = full_agent, Extended = light_agent, Ask = llm_router). For Auto mode conversations, this is the mode selected by the router for this turn. nullable: true enum: - full_agent - light_agent - llm_router generated_title: type: string description: >- Auto-generated conversation title for this turn. Null when title generation does not run or fails. nullable: true tool_calls: type: array description: Array of tool calls that were executed during this response. nullable: true items: allOf: - $ref: '#/components/schemas/ConverseToolCallDelta' - type: object properties: executed_in_parallel: type: boolean description: >- Whether this tool call was executed in parallel with other tool calls. reasoning: type: array description: >- Array of reasoning steps that occurred during this response. Only includes steps with status completed or failed. nullable: true items: $ref: '#/components/schemas/ConverseReasoningDelta' RateLimitError: type: object description: >- Returned when the rate limit is exceeded. Rate limits are enforced per teamspace, endpoint path, and HTTP method over a 60-second sliding window. Each endpoint may have its own limit — check the X-RateLimit-Limit response header for the effective value. required: - error - message - retryable - status_code properties: status_code: type: integer description: The HTTP status code (429). statusCode: type: integer deprecated: true description: Deprecated. Use status_code instead. error: type: string enum: - rate_limit_exceeded description: The error code identifying this as a rate limit error. message: type: string description: A human-readable error message. mitigation: type: string description: Suggested action to resolve the error. retryable: type: boolean description: Whether the request can be retried after a delay. details: type: object properties: reason: type: string description: A detailed explanation of why the rate limit was exceeded. InputItem: type: object title: Input message description: > A message input to the model with a role indicating instruction following `agent` role are presumed to have been generated by the model in previous interactions. properties: type: type: string description: | The type of the message input. Always `message`. enum: - message x-stainless-const: true role: type: string description: | The role of the message input. Always `user`. enum: - user content: description: | Text, file or secret input to the agent. oneOf: - type: string title: Text input description: | A text input to the agent. - $ref: '#/components/schemas/InputMessageContentList' required: - role - content InputAgentConfig: allOf: - $ref: '#/components/schemas/AgentInputConfigCommon' - type: object properties: mcp_servers: type: array items: $ref: '#/components/schemas/AgentMcpServerConfig' description: Registered MCP servers to enable for this agent. nullable: true AgentTools: description: > Array of the agent tools to enable. If not provided, or null is provided - default tools of the agent are used. If empty list provided - none of the tools are used. When connection_id is set for a tool, it will use that specific connection instead of the default one. **Structured outputs (Converse):** All `agent_model` values support JSON Schema constrained responses via **`text.format`** on the Converse request. **Tool availability by agent model** (aligned with in-app **Execute** / **Extended** / direct LLM; see Converse `chat_mode` for **Ask** vs `agent_model`): - **Execute** (`magpie-2.0`, `magpie-2.5`, `magpie-1.1`): All tools below are available. - **Extended** (`magpie-1.1-flash`): Only `semantic_search` is supported. Requests specifying other tools will be rejected with a 400 error. - **Direct LLM** (`llm-only`): No tools are executed. Requests specifying tools will be rejected with a 400 error. Knowledge management tools: - data_analysis: Answer statistical or analytical questions like "Show my quarterly revenue growth" - semantic_search: Search knowledge through natural language queries. - agent_memory: Agents can remember experiences, conversations and user preferences. - schema_info: Helps the Agent understand column names and dataset purpose. Avoid disabling - table_info: Allow the AI Agent to get information about datasets and schemas - create_dataset: Agents respond with data tables - charts: Agents render charts inline in the conversation (bar, line, area, combo, pie, funnel, gauge, treemap, scorecard). Actions: - calendar: Allow the Agent to access and make changes to your Google Calendar - schedule_recurring_message_tool: Eliminate busywork such as: "Send a summary of today's meetings at 5pm on workdays" Data processing tools: - data_classification: Agents handle queries like "Label these emails as high, medium, or low priority" - data_extraction: Helps the agent understand data from other tools. Avoid disabling - image_detection: Extract information from images using AI - pdf_extraction: Extraction of information from PDFs using AI Enhanced response tools: - connect_data: Agents provide buttons to import data in response to queries like "Connect Hubspot" - download_data: Agents handle queries like "download the table as CSV" Web tools: - web_search: Agents search the internet, and provide links to their sources - fetch_url: Fetch URL content - company_prospect_researcher: Agents provide information about companies - people_prospect_researcher: Agents provide information about people type: array nullable: true items: oneOf: - $ref: '#/components/schemas/ToolName' - $ref: '#/components/schemas/Tool' InputMcpServer: type: object description: > Configuration for an MCP (Model Context Protocol) server passed directly in the request. MCP servers provide additional tools that extend the agent's capabilities. Servers must implement the MCP streamable HTTP transport, including `initialize`, `notifications/initialized`, `tools/list`, and `tools/call`. required: - type - server_label - server_url properties: type: type: string enum: - inline_mcp description: >- The type of MCP server configuration. Use 'inline_mcp' for server configs passed directly in the request. x-stainless-const: true server_label: type: string description: >- A unique label to identify this MCP server. Used for tool namespacing. maxLength: 64 server_url: type: string format: uri description: The HTTPS URL of the MCP streamable HTTP endpoint. server_description: type: string description: Optional description of what this MCP server provides. nullable: true maxLength: 500 authorization: type: string description: >- Value sent in the `Authorization` header when calling the MCP server (e.g., 'Bearer '). nullable: true AgentRouting: type: object description: Determines how the API routes the converse request to an agent. discriminator: propertyName: mode oneOf: - type: object title: Auto routing description: >- The API automatically selects the best agent from the entire Teamspace. required: - mode properties: mode: type: string enum: - auto description: The API selects the best agent from the entire Teamspace. - type: object title: Manual routing description: The API selects the best agent from the specific list you provide. required: - mode - targets properties: mode: type: string enum: - manual description: >- The API selects the best agent from the specific list you provide. targets: type: array minItems: 1 description: >- Limit the selection pool to these specific agents. Each item may be an agent ID string or an inline agent config object. items: oneOf: - type: string description: Agent ID string - $ref: '#/components/schemas/AgentConfigWithId' UserOverride: type: object description: >- User information override for converse calls. All fields are optional - only provided fields will override the default user information. properties: first_name: type: string description: Override the user's first name for this converse call. nullable: true last_name: type: string description: Override the user's last name for this converse call. nullable: true email: type: string description: Override the user's email for this converse call. nullable: true ConversationMessage: type: object description: >- The `conversation.message` object represents a message in a conversation. required: - object - id - agent_id - role - content - created_at - conversation_id - credits - citations properties: object: type: string enum: - conversation.message description: The object type, which is always `conversation.message`. id: type: string description: The message identifier. agent_id: type: string description: The ID of the agent that sent or responded to the message. role: type: string enum: - user - agent description: The role of the message sender - either 'user' or 'agent'. content: type: array description: Contents of the message. items: oneOf: - $ref: '#/components/schemas/MessageContentText' - $ref: '#/components/schemas/MessageContentVoice' - $ref: '#/components/schemas/MessageContentFile' discriminator: propertyName: type mapping: text: $ref: '#/components/schemas/MessageContentText' voice: $ref: '#/components/schemas/MessageContentVoice' input_file: $ref: '#/components/schemas/MessageContentFile' created_at: type: string format: date-time description: The ISO string for when the message was created. conversation_id: type: string description: The ID of the conversation the message belongs to. credits: nullable: true description: >- Credit consumption for this converse turn. `null` for user-role messages and when retrieving messages from conversation history. allOf: - $ref: '#/components/schemas/OperationCredits' citations: $ref: '#/components/schemas/Citations' nullable: true ConverseToolCallDelta: type: object description: > Event emitted when tools are called during the agent's reasoning process. required: - type - id - status - tool properties: type: type: string enum: - tool_call id: type: string description: The ID of the tool call. status: type: string enum: - in_progress - completed - failed tool: $ref: '#/components/schemas/ToolDef' output: type: string description: The output of the tool call. ConverseReasoningDelta: type: object required: - id - status - type properties: id: type: string description: The ID of the reasoning step. status: type: string enum: - in_progress - completed - failed description: The current status of the reasoning step. type: type: string enum: - reasoning description: The type of the delta, which is always reasoning. task: type: string description: The task description for this reasoning step. nullable: true output: type: string description: >- The output of the reasoning step. Only present when status is completed or failed. nullable: true execution_time_ms: type: number description: The execution time of the reasoning step in milliseconds. nullable: true executed_in_parallel: type: boolean description: >- Whether this reasoning step was executed in parallel with other steps. nullable: true InputMessageContentList: type: array title: Input item content list description: > A list of one or many input items to the model, containing different content types. items: $ref: '#/components/schemas/InputContent' AgentInputConfigCommon: allOf: - $ref: '#/components/schemas/AgentConfigShared' - type: object properties: tools: $ref: '#/components/schemas/AgentTools' nullable: true disabled_tools: $ref: '#/components/schemas/AgentTools' description: >- Array of the agent tools to disable. Disabling is performed after the 'agent_tools' rules are applied. For example, agent_tools: null and disabled_tools: [data_analysis] will enable everything but the data_analysis tool. If nothing or [] is provided, nothing is disabled and therefore only the agent_tools setting is relevant. nullable: true llm_model: $ref: '#/components/schemas/LlmModel' nullable: true temperature: type: number description: >- Sampling temperature for model output. Lower values are more deterministic; higher values are more diverse. minimum: 0 maximum: 1 nullable: true agent_model: $ref: '#/components/schemas/AgentModel' nullable: true AgentMcpServerConfig: type: object required: - server_id properties: server_id: type: string credential_id: type: string nullable: true ToolName: type: string description: The unique identifier for a tool. Tool: type: object required: - name properties: name: $ref: '#/components/schemas/ToolName' connection_id: type: string description: The ID of the connection to use for the tool. nullable: true AgentConfigWithId: allOf: - type: object properties: agent_id: type: string description: The ID of the agent to use for routing. - $ref: '#/components/schemas/InputAgentConfig' MessageContentText: type: object description: Text content for a message. required: - type - text properties: type: type: string enum: - text x-stainless-const: true text: type: string description: The text content of the message. MessageContentVoice: type: object description: Voice content for a message. required: - type - audio_clips - transcript - duration_ms properties: type: type: string enum: - voice x-stainless-const: true audio_clips: type: array description: Array of audio clips with timestamps for synchronized playback. items: $ref: '#/components/schemas/VoiceAudioClip' transcript: type: string description: User transcript of the voice message. agent_transcript: type: string description: Agent transcript of the voice message (for agent role messages). nullable: true duration_ms: type: number description: Total duration of the voice message in milliseconds. nullable: true timeline_events: type: array description: >- Per-turn transcript, citation, and tool-call status events in chronological order. Each entry includes a timestamp offset from the start of the voice session. Present only for voice sessions that recorded timeline data. items: $ref: '#/components/schemas/VoiceTimelineEvent' MessageContentFile: type: object description: >- File attachment content for a message. Represents a file that was uploaded as part of the user's message. required: - type - file_id properties: type: type: string enum: - input_file x-stainless-const: true file_id: type: string description: >- The ID of the attached file. Use this ID with the files API to download the file content. OperationCredits: type: object required: - consumed properties: consumed: type: number description: The number of credits consumed by the operation. Citations: type: array description: >- Array of citations that provide sources for factual statements in the response. Each citation includes the referenced text and its sources. items: type: object required: - citation - sources properties: citation: type: string description: The text snippet from the response that is being cited. sources: type: array description: Array of sources that support this citation. items: $ref: '#/components/schemas/CitationSource' ToolDef: type: object description: The `Tool` object represents a tool that can be used by agents. required: - object - name - label - description properties: object: type: string enum: - tool description: The object type, which is always `tool`. name: type: string description: The unique identifier name used to reference the tool. label: type: string description: The display name of the tool. description: type: string description: A detailed description of what the tool does. InputContent: oneOf: - $ref: '#/components/schemas/InputText' - $ref: '#/components/schemas/InputFile' - $ref: '#/components/schemas/InputSecret' - $ref: '#/components/schemas/InputKnowledge' - $ref: '#/components/schemas/InputPage' AgentConfigShared: type: object properties: knowledge_ids: type: array nullable: true deprecated: true description: >- Deprecated, use corpus instead. Array of Knowledge IDs the agent should use during the converse. When omitted, all knowledge is used. items: type: string corpus: type: array nullable: true description: >- Array of corpus items the agent should use during the converse. When omitted, all knowledge is used. items: oneOf: - $ref: '#/components/schemas/CorpusKnowledgeItem' - $ref: '#/components/schemas/CorpusPageItem' system_prompt: type: string description: Directs your AI Agent's operational behavior. nullable: true custom_prompt: type: string description: >- Use custom prompt to instruct the style and formatting of the agent's response nullable: true planning_prompt: type: string description: >- Define the planning strategy your AI Agent should use when breaking down tasks and solving problems nullable: true LlmModel: anyOf: - $ref: '#/components/schemas/LlmModelEnum' - type: string default: gemini-2.5-flash description: >- The LLM used to generate responses. Deprecated Gemini 2.0 Flash ids are accepted for backward compatibility and automatically run as gemini-3.1-flash-lite. AgentModel: anyOf: - $ref: '#/components/schemas/AgentModelEnum' - type: string type: string enum: - magpie-1.1 - magpie-1.1-flash - magpie-2.0 - magpie-2.5 - magpie-2.5-flash - llm-only default: magpie-2.0 description: > The agent model determines the processing mode for Converse requests. The Datagrid web app exposes **Ask**, **Extended**, and **Execute** as Converse **`chat_mode`** (`llm_router`, `light_agent`, `full_agent`). The values below set **`config.agent_model`** (model tier and tool limits)—use both fields when mirroring in-app behavior. **Execute** (full tool use, planning, and multi-step reasoning; aligns with **Execute** in the web app / `full_agent`): - `magpie-2.0` — Default. Full agent model with proactive planning and reasoning. - `magpie-2.5` — Beta. Latest full-agent model — faster, more adaptable, and built to handle a broader range of real-world tasks. - `magpie-1.1` — Previous-generation full agent model. **Extended** (search-focused; aligns with **Extended** in the web app / `light_agent`; not **Ask**): - `magpie-1.1-flash` — Fast model optimized for RAG use cases. Only supports the `semantic_search` tool. A 400 error will be returned if other tools are specified. **Direct LLM** (no tool execution; **`agent_model` only**—**Ask** in the web app is `chat_mode: llm_router`, not `magpie-1.1-flash`): - `llm-only` — Runs a direct LLM conversation with no planning or tool calls. A 400 error will be returned if tools are specified. Can also accept any custom string value for future model versions. nullable: true VoiceAudioClip: type: object description: A single audio clip from a voice message. required: - id - audio_uri - start_time_ms - duration_ms - participant properties: id: type: string description: >- Unique identifier for the audio clip file. Use this ID with the files API to download the audio content. audio_uri: type: string description: Datagrid file URI for the audio file (WAV format). start_time_ms: type: number description: >- Start time of this clip relative to the beginning of the voice message, in milliseconds. duration_ms: type: number description: Duration of this audio clip in milliseconds. participant: type: object description: Participant who spoke in this clip. required: - type - id properties: type: type: string enum: - user - agent id: type: string VoiceTimelineEvent: type: object description: >- A single event from a voice session timeline, representing a transcript turn, citation, or tool-call status update. required: - type - timestamp_ms properties: type: type: string enum: - transcript - citation - tool_call_status description: The type of timeline event. timestamp_ms: type: number description: >- Timestamp offset from the start of the voice session, in milliseconds. role: type: string enum: - user - agent description: The role of the participant for this event. text: type: string description: >- Plain text transcript for this turn. Present when type is 'transcript'. citations: $ref: '#/components/schemas/Citations' description: Citations for this event. Present when type is 'citation'. tool_name: type: string description: >- Name of the tool whose status changed. Present when type is 'tool_call_status'. status: type: string enum: - started - completed - failed description: >- Tool-call status for this event. Present when type is 'tool_call_status'. CitationSource: type: object required: - type - source_name - confirmations properties: type: type: string enum: - image - pdf_page - record - web_search - sql_query_result - action source_id: type: string description: Id of the source. source_name: type: string description: Name of the source. source_uri: type: string description: URI of the source. confirmations: type: array description: An array of text snippets from the source that confirm the citation. items: type: string InputText: type: object title: Text input description: | A text input to the model. properties: type: type: string description: | The type of the input item. Always `input_text`. enum: - input_text x-stainless-const: true text: type: string description: | The text input to the model. required: - type - text InputFile: type: object title: File input description: | A file input to the model. properties: type: type: string description: | The type of the input item. Always `input_file`. enum: - input_file x-stainless-const: true file_id: type: string description: | The ID of the file to be sent to the model. required: - type - file_id InputSecret: type: object title: Secret input description: | A secret input to the model. properties: type: type: string description: | The type of the input item. Always `input_secret`. enum: - input_secret x-stainless-const: true secret_id: type: string description: | The ID of the secret to be sent to the model. required: - type - secret_id InputKnowledge: type: object title: Knowledge input description: > A knowledge reference input to the model. This references knowledge by ID. The knowledge will be made accessible to the agent, and will be included in the prompt provided to the agent. The position of this reference relative to other text of the input impact the agent's interpretation. properties: type: type: string description: | The type of the input item. Always `input_knowledge`. enum: - input_knowledge x-stainless-const: true knowledge_id: type: string description: | The ID of the knowledge to be referenced. required: - type - knowledge_id InputPage: type: object title: Page input description: > A page reference input to the model. This references a page by ID. The page, and all knowledge under it, will be made accessible to the agent, and a reference to the page will be included in the prompt provided to the agent. The position of this reference relative to other text of the input will impact the agent's interpretation. properties: type: type: string description: | The type of the input item. Always `input_page`. enum: - input_page x-stainless-const: true page_id: type: string description: | The ID of the page to be referenced. required: - type - page_id CorpusKnowledgeItem: type: object required: - type - knowledge_id properties: type: type: string enum: - knowledge description: The type of the corpus item. Always 'knowledge' for knowledge items. x-stainless-const: true knowledge_id: type: string description: The ID of the knowledge to include in the corpus. CorpusPageItem: type: object required: - type - page_id properties: type: type: string enum: - page description: The type of the corpus item. Always 'page' for page items. x-stainless-const: true page_id: type: string description: The ID of the page to include in the corpus. LlmModelEnum: type: string description: >- The LLM used to generate responses. Deprecated Gemini 2.0 Flash ids are accepted for backward compatibility and automatically run as gemini-3.1-flash-lite. enum: - gemini-3.5-flash - gemini-3.1-flash-lite - gemini-3-pro-preview - gemini-3.1-pro-preview - gemini-3-flash-preview - gemini-2.5-pro - gemini-2.5-pro-preview-05-06 - gemini-2.5-flash - gemini-2.5-flash-preview-04-17 - gemini-2.5-flash-native-audio-preview-12-2025 - gemini-2.5-flash-lite - gpt-5 - gpt-5.1 - gemini-2.0-flash-001 - gemini-2.0-flash - gemini-1.5-pro-001 - gemini-1.5-pro-002 - gemini-1.5-flash-002 - gemini-1.5-flash-001 - chatgpt-4o-latest - gpt-4o - gpt-4 - gpt-4-turbo - gpt-4o-mini - claude-haiku-4-5@20251001 - claude-sonnet-4-6@default - claude-opus-4-6@default - claude-opus-4-7 - claude-opus-4-8 - anthropic.claude-haiku-4-5-20251001-v1:0 - anthropic.claude-sonnet-4-6 - anthropic.claude-opus-4-6-v1 - anthropic.claude-opus-4-7 - anthropic.claude-opus-4-8 AgentModelEnum: type: string enum: - magpie-1.1 - magpie-1.1-flash - magpie-2.0 - magpie-2.5 - magpie-2.5-flash - llm-only description: Predefined agent model versions securitySchemes: BearerAuth: type: http scheme: bearer ````