> ## Documentation Index > Fetch the complete documentation index at: https://docs.notte.cc/llms.txt > Use this file to discover all available pages before exploring further. # Agent Start ## OpenAPI ````yaml https://api.notte.cc/openapi.json post /agents/start openapi: 3.1.0 info: title: Notte API description: >- Notte API is a REST API that allows you to interact with Notte. It is used to create cloud browser sessions, scrape webpages, and run web ai agents to act on your behalf on the internet. version: 1.4.40 x-logo: url: https://www.notte.cc/images/logo/logo-white.png servers: [] security: [] tags: - name: agents description: Web AI agents (start, stop, status, replay, etc.) - name: sessions description: Session management (start, stop, status, etc.) - name: debug description: Session debugging tools (replay,logs, recording, etc.) - name: page description: Page operations withing a session (observe, step, scrape, etc.) - name: storage description: File storage interface (upload, download, list, etc.) - name: network description: Network requests/responses withing a session (intercept, etc.) - name: vaults description: >- Vault & Credentials management (create/delete vaults, create/delete credentials, etc.) - name: personas description: Persona management (create, delete, list emails, list sms, etc.) - name: scrape description: >- Webpage scraping (scrape, screenshot, etc.) with automatic session management. - name: health description: Health check endpoint. - name: usage description: Usage logs (usage, logs, etc.) - name: functions description: Functions management (create, delete, list, etc.) paths: /agents/start: post: tags: - agents summary: Agent Start operationId: agent_start parameters: - name: x-notte-request-origin in: header required: false schema: anyOf: - type: string - type: 'null' title: X-Notte-Request-Origin - name: x-notte-sdk-version in: header required: false schema: anyOf: - type: string - type: 'null' title: X-Notte-Sdk-Version requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/ApiAgentStartRequest' responses: '200': description: Successful Response content: application/json: schema: $ref: '#/components/schemas/AgentResponse' '422': description: Validation Error content: application/json: schema: $ref: '#/components/schemas/HTTPValidationError' security: - OAuth2PasswordBearer: [] x-codeSamples: - lang: Python source: |- from notte_sdk import NotteClient notte = NotteClient() task = "find a cute dog on google images" with notte.Session() as session: agent = notte.Agent(session=session, max_steps=5) # this runs the agent and waits for completion full_response = agent.run(task=task) print(f"{full_response.success=}, {full_response.answer=}") label: python - lang: JavaScript source: |- import { NotteClient } from "notte-sdk"; const notte = new NotteClient(); const task = "find a cute dog on google images"; await notte.Session().use(async (session) => { const agent = notte.Agent({ session, max_steps: 5 }); // this runs the agent and waits for completion const response = await agent.run({ task }); console.log(`success=${response.success}, answer=${response.answer}`); }); label: node - lang: Curl source: |- curl -X POST "https://api.notte.cc/agents/start" \ -H "Authorization: Bearer $NOTTE_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "task": "task", "url": null, "response_format": null, "session_offset": null, "reasoning_model": "vertex_ai/gemini-2.5-flash", "use_vision": true, "max_steps": 20, "vault_id": null, "persona_id": null, "notifier_config": null, "session_id": "" }' label: curl components: schemas: ApiAgentStartRequest: properties: task: type: string title: Task description: The task that the agent should perform url: anyOf: - type: string - type: 'null' title: Url description: The URL that the agent should start on (optional) response_format: anyOf: - {} - type: 'null' title: Response Format description: >- The response format to use for the agent answer. You can use a Pydantic model or a JSON Schema dict (cf. https://docs.pydantic.dev/latest/concepts/json_schema/#generating-json-schema.) session_offset: anyOf: - type: integer - type: 'null' title: Session Offset description: >- [Experimental] The step from which the agent should gather information from in the session. If none, fresh memory reasoning_model: anyOf: - $ref: '#/components/schemas/LlmModel' - type: string title: Reasoning Model description: The reasoning model to use use_vision: type: boolean title: Use Vision description: >- Whether to use vision for the agent. Not all reasoning models support vision. default: true max_steps: type: integer maximum: 150 minimum: 1 title: Max Steps description: The maximum number of steps the agent should take default: 20 vault_id: anyOf: - type: string - type: 'null' title: Vault Id description: The vault to use for the agent persona_id: anyOf: - type: string - type: 'null' title: Persona Id description: The persona to use for the agent notifier_config: anyOf: - additionalProperties: true type: object - type: 'null' title: Notifier Config description: Config used for the notifier session_id: type: string title: Session Id description: The ID of the session to run the agent on additionalProperties: false type: object required: - task - session_id title: ApiAgentStartRequest AgentResponse: properties: agent_id: type: string title: Agent Id description: The ID of the agent created_at: type: string format: date-time title: Created At description: The creation time of the agent session_id: type: string title: Session Id description: The ID of the session status: $ref: '#/components/schemas/AgentStatus' description: The status of the agent (active or closed) closed_at: anyOf: - type: string format: date-time - type: 'null' title: Closed At description: The closing time of the agent saved: type: boolean title: Saved description: Whether the agent is saved as a workflow default: false type: object required: - agent_id - created_at - session_id - status title: AgentResponse HTTPValidationError: properties: detail: items: $ref: '#/components/schemas/ValidationError' type: array title: Detail type: object title: HTTPValidationError LlmModel: type: string enum: - openai/gpt-4o - gemini/gemini-2.5-flash - vertex_ai/gemini-2.5-flash - openrouter/google/gemma-3-27b-it - cerebras/gpt-oss-120b - groq/gpt-oss-120b - perplexity/sonar-pro - deepseek/deepseek-r1 - together_ai/meta-llama/llama-3.3-70b-instruct - anthropic/claude-sonnet-4-5-20250929 - moonshot/kimi-k2.5 - xai/grok-4-1-fast-non-reasoning - minimax/minimax-m2.5 title: LlmModel AgentStatus: type: string enum: - active - closed title: AgentStatus ValidationError: properties: loc: items: anyOf: - type: string - type: integer type: array title: Location msg: type: string title: Message type: type: string title: Error Type input: title: Input ctx: type: object title: Context type: object required: - loc - msg - type title: ValidationError securitySchemes: OAuth2PasswordBearer: type: oauth2 flows: password: scopes: {} tokenUrl: token ````