> ## Documentation Index > Fetch the complete documentation index at: https://docs.devic.ai/llms.txt > Use this file to discover all available pages before exploring further. # Continue a thread with a user message > Appends a user message to an existing thread to continue it. The effect depends on the current thread state: a completed/failed/waiting thread is re-queued and re-run with the message appended; a still-running or queued thread receives the message on its next turn (right after the pending tool response). Threads waiting for approval, handed off to a subagent, or still being created cannot be continued this way. ## OpenAPI ````yaml POST /v1/agents/threads/{threadId}/messages openapi: 3.0.0 info: title: Devic.ai Public API description: >- Devic.ai is an AI platform that allows you to create, manage, and use AI agents for various tasks. version: 1.0.0 contact: name: Devic.ai Support url: https://devic.ai x-logo: url: https://devic.ai/logo.png altText: Devic.ai Logo x-summary: Public API for interacting with Devic.ai platform servers: - url: https://api.devic.ai description: Production server - url: https://staging-api.devic.ai description: Staging server security: - bearerAuth: [] tags: - name: Projects description: Group agents, assistants, documents and costs into projects - name: Documents description: >- Knowledge base documents: create, version, attach and index markdown content for RAG - name: Document Folders description: Organise knowledge base documents into folders and attach them in bulk - name: Files description: Upload files and obtain shareable download URLs to attach to messages - name: Agents description: Endpoints related to AI agents and their operations - name: Assistants description: Endpoints for interacting with assistants and their specializations - name: Tool Servers description: Endpoints for managing tool servers and their tool definitions - name: Health description: API health check endpoints - name: Documentation description: Endpoints for retrieving markdown documentation - name: Integrations description: Connect third-party apps and turn them into tools - name: Triggers description: Start an agent or an assistant from an app event - name: Tenant Integrations description: Apps that each end user connects for themselves - name: Memory description: What an assistant remembers between conversations - name: Skills description: Reusable instruction packs for agents and assistants - name: Speech to Text description: Audio transcription - name: Tenants description: Tenants, subtenants and their usage - name: MCP Gateway description: One MCP endpoint over many servers, with visibility per user - name: Tenant Sessions description: Tokens that prove which end user is calling paths: /v1/agents/threads/{threadId}/messages: post: tags: - Agents summary: Continue a thread with a user message description: >- Appends a user message to an existing thread to continue it. The effect depends on the current thread state: a completed/failed/waiting thread is re-queued and re-run with the message appended; a still-running or queued thread receives the message on its next turn (right after the pending tool response). Threads waiting for approval, handed off to a subagent, or still being created cannot be continued this way. operationId: addThreadMessage parameters: - name: threadId required: true in: path description: The ID of the thread to continue schema: type: string - name: devic-api-source required: false in: header schema: type: string description: >- Free-text label recorded with the call, to tell apart the integrations that use the same key. requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/AddThreadMessageDto' responses: '200': description: Message added to the thread content: application/json: schema: $ref: '#/components/schemas/ThreadResponseDto' '401': description: Unauthorized - Invalid or missing token '404': description: Thread not found '409': description: Thread is in a state that cannot be continued with a message '429': description: Too Many Requests - Rate limit exceeded '500': description: Internal Server Error security: - bearerAuth: [] components: schemas: AddThreadMessageDto: type: object properties: message: description: >- The user message to append to the thread to continue it (string or object). Behaviour depends on the current thread state: a running/queued thread receives it on its next turn (after the pending tool response); a completed/failed/waiting thread is re-queued to run again with the message appended. example: Also include a summary of the previous steps oneOf: - type: string - type: object required: - message ThreadResponseDto: type: object properties: tokenUsage: type: object description: Enhanced token usage with cost breakdown properties: inputTokens: type: number example: 1250 outputTokens: type: number example: 450 inputCachedTokens: type: number example: 320 outputCachedTokens: type: number example: 80 totalTokens: type: number example: 1700 totalCachedTokens: type: number example: 400 cost: type: object properties: inputCost: type: number example: 0.00125 outputCost: type: number example: 0.0045 inputCachedCost: type: number example: 0.00016 outputCachedCost: type: number example: 0.008 totalCost: type: number example: 0.00641 provider: type: string example: open-ai model: type: string example: gpt-4o-mini-2024-07-18 metadata: type: object description: >- Thread metadata containing prompt template parameters and tenant token properties: propmptTemplateParams: type: object description: >- Parameters used for template variable substitution in the agent system prompt (e.g., {{customerName}} becomes "John Doe") tenantToken: type: string description: Token used for authentication in tool calling operations tenantId: type: string description: Tenant ID for multi-tenant environments example: tenant-123 isSubthread: type: boolean description: Whether this thread is a subthread created by a parent handoff example: false parentThreadId: type: string description: ID of the parent thread (if this is a subthread) example: 64f5a5b8c123456789abcdef parentAgentId: type: string description: ID of the parent agent (if this is a subthread) example: 64f5a5b8c123456789abcdef subThreadToolCallId: type: string description: Tool call ID in the parent thread that triggered this subthread example: call_abc123 pendingHandOffSubThreadIds: description: IDs of subthreads still executing (for parent threads) type: array items: type: string handOffSubThreadIds: description: All subthread IDs ever created during handoffs (for parent threads) type: array items: type: string tasks: description: List of task for the thread execution type: array items: $ref: '#/components/schemas/TaskDto' pendingAsyncToolCalls: type: array description: >- Pending async tool calls awaiting an external response. Each entry exposes the callbackUrl the external system must POST the response to. items: type: object properties: toolCallId: type: string toolName: type: string sentAt: type: number resolved: type: boolean callbackUrl: type: string pausedReason: type: string description: >- Reason the thread is paused (e.g. waiting for an async tool response). pausedUntil: type: number description: >- Epoch milliseconds when a paused thread is scheduled to resume. Only set while the thread is paused with a scheduled resume. example: 1721551200000 pausedTimestampMS: type: number description: Epoch milliseconds when the thread was paused. example: 1721550000000 coreMemories: description: >- Audit trail of the core-memory blocks this thread saw: one snapshot (render revision + the injected entries as structured JSON) per distinct revision, including mid-run edits. Present only for agents with core memory enabled. type: array items: type: string recalledMemories: description: >- Structured recall events of this thread: the automatic long-term-memory recall at run start plus every memory-tool lookup, each with the query and the facts/entities it surfaced. Present only for agents with long-term memory enabled. type: array items: type: string TaskDto: type: object properties: {} securitySchemes: bearerAuth: type: http scheme: bearer bearerFormat: JWT description: Use JWT token for authentication ````