> ## Documentation Index > Fetch the complete documentation index at: https://www.krea.ai/docs/llms.txt > Use this file to discover all available pages before exploring further. # Send Agent Message > Start a Krea Agent session or send a follow-up turn by passing sessionId. Returns immediately while the Agent works in the background. Sessions are visible at the returned url and billed to the workspace's compute unit balance. The Agent does not ask questions, and cost and permission prompts are approved automatically: include all requirements in the prompt. A session runs one turn at a time. Attachments accept http(s) URLs and authorized Krea Files URIs (file:). Model pins must be included in the workspace plan. ## OpenAPI ````yaml https://api.krea.ai/openapi.json post /agent/send-message openapi: 3.1.0 info: title: Krea API version: v1 servers: - url: https://api.krea.ai description: Krea API security: - bearerAuth: [] tags: - name: General description: Core API operations including job management and billing information - name: Krea description: Krea-owned generation models including Krea 2 - name: Assets description: >- Asset management endpoints for uploading and managing images, videos, audio files, and 3D models - name: Styles description: Style (LoRA) generation and management endpoints - name: Node Apps description: Executing custom node apps built in our nodes tool - name: Image description: Image generation endpoints - name: Image Enhance description: Image enhance endpoints - name: Video description: Video generation endpoints - name: Video Enhance description: Video enhance and upscale endpoints - name: 3D description: 3D model generation endpoints - name: Audio description: Speech and audio generation endpoints - name: Usage description: Workspace usage reporting - name: Agent description: Krea Agent sessions - name: Utilities description: Miscellaneous utility endpoints paths: /agent/send-message: post: tags: - Agent summary: Send Agent Message description: >- Start a Krea Agent session or send a follow-up turn by passing sessionId. Returns immediately while the Agent works in the background. Sessions are visible at the returned url and billed to the workspace's compute unit balance. The Agent does not ask questions, and cost and permission prompts are approved automatically: include all requirements in the prompt. A session runs one turn at a time. Attachments accept http(s) URLs and authorized Krea Files URIs (file:). Model pins must be included in the workspace plan. requestBody: required: true content: application/json: schema: type: object properties: sessionId: description: >- Omit to start a new session; pass an existing id to continue it type: string format: uuid pattern: >- ^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$ prompt: default: '' type: string maxLength: 100000 attachments: default: [] description: >- Transfer reference files that are not otherwise accessible to the Agent here. Usable URLs in the prompt, existing Krea context, and prior session files need not be repeated in attachments. The Agent does not inherit your conversation or attachments. Mentioning OpenAI file_ IDs in prompt text does not transfer their bytes, and those IDs cannot be converted to Krea file URIs. maxItems: 20 type: array items: type: object properties: url: anyOf: - type: string format: uri - type: string pattern: >- ^(file):([0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12})$ description: >- An accessible http(s) URL, or an authorized Krea Files URI "file:" from list_files or write_files. OpenAI file_ IDs and caller-local paths are not supported; transfer the actual file first. name: type: string minLength: 1 maxLength: 255 contentType: type: string minLength: 1 maxLength: 255 required: - url model: description: >- Pin an LLM. Omit for Auto. Must be included in the workspace plan. type: string enum: - openai/gpt-6-astra - openai/gpt-6.1-sol - openai/gpt-6-sol - openai/gpt-6-luna - openai/gpt-5.6-sol - openai/gpt-5.6-terra - openai/gpt-5.6-luna - openai/gpt-5.5 - anthropic/claude-opus-5.5 - anthropic/claude-opus-5 - anthropic/claude-fable-5.1 - anthropic/claude-fable-5 - anthropic/claude-opus-4.8 - anthropic/claude-sonnet-5 - anthropic/claude-sonnet-4.5 - anthropic/claude-opus-4.5 - google/gemini-4-argon - google/gemini-3.8-flash - google/gemini-3.5-flash - google/gemini-3.5-flash-lite - google/gemini-3.1-pro-preview - google/gemini-3.1-flash-lite - google/gemini-2.5-flash-lite - xai/grok-4.5 - zai/glm-5.2 - moonshotai/kimi-k3 effort: type: string enum: - fast - default - high - xhigh name: description: Session name, new sessions only type: string minLength: 1 maxLength: 120 desktopAppId: description: >- An appId from list_desktop_apps. The Agent then works inside that desktop app for this turn. type: string format: uuid pattern: >- ^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$ responses: '200': description: The Agent session result. content: application/json: schema: type: object properties: sessionId: type: string threadId: type: string runId: type: string status: type: string enum: - running url: type: string required: - sessionId - threadId - runId - status - url '400': description: Invalid JSON body. content: application/json: schema: type: object properties: error: type: string code: type: string allowedModels: type: array items: type: string required: - error additionalProperties: false '401': description: Authentication required. content: application/json: schema: type: object properties: error: type: string code: type: string allowedModels: type: array items: type: string required: - error additionalProperties: false '402': description: Insufficient compute unit balance or free message limit reached. content: application/json: schema: type: object properties: error: type: string code: type: string allowedModels: type: array items: type: string required: - error additionalProperties: false '403': description: The requested model is not included in the workspace plan. content: application/json: schema: type: object properties: error: type: string code: type: string allowedModels: type: array items: type: string required: - error additionalProperties: false '404': description: Session or attachment not found or not accessible. content: application/json: schema: type: object properties: error: type: string code: type: string allowedModels: type: array items: type: string required: - error additionalProperties: false '409': description: A turn is already in progress. content: application/json: schema: type: object properties: error: type: string code: type: string allowedModels: type: array items: type: string required: - error additionalProperties: false '422': description: Invalid input or desktop app unavailable. content: application/json: schema: type: object properties: error: type: string code: type: string allowedModels: type: array items: type: string required: - error additionalProperties: false components: securitySchemes: bearerAuth: scheme: bearer bearerFormat: Bearer type: http ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.