openapi: 3.0.3
info:
  title: AgentQL API
  description: AgentQL connects LLMs and AI agents to the entire web through natural language queries, enabling structured data extraction from web pages, documents, and browser sessions. Generated from documentation.
  version: v1
  x-generated-from: documentation
servers:
  - url: https://api.agentql.com/v1
    description: AgentQL Production API

security:
  - ApiKeyAuth: []

paths:
  /query-data:
    post:
      operationId: queryWebPageData
      summary: AgentQL Query Web Page Data
      description: Extract structured JSON data from a web page using AgentQL query language or natural language prompt. Supports URL-based and raw HTML input with configurable browser behavior.
      tags:
        - Query Data
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/QueryDataRequest'
            examples:
              QueryDataRequest200Example:
                summary: Default queryWebPageData request
                x-microcks-default: true
                value:
                  url: "https://news.ycombinator.com"
                  query: "{ posts[] { title link score } }"
                  params:
                    mode: standard
                    wait_for: 2000
      responses:
        '200':
          description: Successful data extraction
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/QueryDataResponse'
              examples:
                QueryWebPageData200Example:
                  summary: Default queryWebPageData 200 response
                  x-microcks-default: true
                  value:
                    data:
                      posts:
                        - title: "Show HN: AgentQL connects AI agents to the web"
                          link: "https://news.ycombinator.com/item?id=500123"
                          score: 142
                    metadata:
                      request_id: "req-a1b2c3d4"
                      response_time_ms: 1847
        '400':
          description: Bad request — invalid query or missing required parameters
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ErrorResponse'
        '401':
          description: Unauthorized — invalid or missing API key
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ErrorResponse'
        '429':
          description: Rate limit exceeded
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ErrorResponse'
      x-microcks-operation:
        delay: 0
        dispatcher: FALLBACK

  /tetra/sessions:
    post:
      operationId: createRemoteBrowserSession
      summary: AgentQL Create Remote Browser Session
      description: Establish a managed remote Chrome browser session with Chrome DevTools Protocol (CDP) access for authenticated web automation and stealth browsing.
      tags:
        - Remote Browser
      requestBody:
        required: false
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/CreateSessionRequest'
            examples:
              CreateRemoteBrowserSessionRequestExample:
                summary: Default createRemoteBrowserSession request
                x-microcks-default: true
                value:
                  browser_ua_preset: macos
                  browser_profile: stealth
                  inactivity_timeout_seconds: 300
      responses:
        '200':
          description: Browser session created successfully
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/CreateSessionResponse'
              examples:
                CreateRemoteBrowserSession200Example:
                  summary: Default createRemoteBrowserSession 200 response
                  x-microcks-default: true
                  value:
                    session_id: "sess-a1b2c3d4e5f6"
                    cdp_url: "wss://tetra.agentql.com/sessions/sess-a1b2c3d4e5f6/cdp"
                    expires_at: "2026-04-19T15:30:00Z"
        '401':
          description: Unauthorized
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ErrorResponse'
      x-microcks-operation:
        delay: 0
        dispatcher: FALLBACK

  /query-document:
    post:
      operationId: queryDocument
      summary: AgentQL Query Document
      description: Extract structured data from PDF documents or images (JPEG, PNG) using AgentQL query language or natural language prompts.
      tags:
        - Query Document
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/QueryDocumentRequest'
            examples:
              QueryDocumentRequestExample:
                summary: Default queryDocument request
                x-microcks-default: true
                value:
                  url: "https://example.com/invoice.pdf"
                  query: "{ invoice { total_amount vendor_name date items[] { description amount } } }"
      responses:
        '200':
          description: Successful document extraction
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/QueryDataResponse'
              examples:
                QueryDocument200Example:
                  summary: Default queryDocument 200 response
                  x-microcks-default: true
                  value:
                    data:
                      invoice:
                        total_amount: 4250.00
                        vendor_name: "Acme Corp"
                        date: "2026-04-01"
                        items:
                          - description: "Professional Services"
                            amount: 4000.00
                          - description: "Travel Expenses"
                            amount: 250.00
                    metadata:
                      request_id: "req-d4e5f6a1"
                      response_time_ms: 2103
        '400':
          description: Bad request
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ErrorResponse'
        '401':
          description: Unauthorized
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ErrorResponse'
      x-microcks-operation:
        delay: 0
        dispatcher: FALLBACK

components:
  securitySchemes:
    ApiKeyAuth:
      type: apiKey
      in: header
      name: X-API-Key
      description: API key from the AgentQL Dev Portal (https://dev.agentql.com)

  schemas:
    QueryDataRequest:
      type: object
      description: Request body for querying data from a web page.
      properties:
        url:
          type: string
          format: uri
          description: The URL of the web page to query.
          example: "https://news.ycombinator.com"
        html:
          type: string
          description: Raw HTML content to query instead of a URL.
          example: "<html><body><h1>Hello World</h1></body></html>"
        query:
          type: string
          description: AgentQL query string for structured data extraction.
          example: "{ posts[] { title link score } }"
        prompt:
          type: string
          description: Natural language description of the data to extract (alternative to query).
          example: "Extract the top 10 posts with their titles, links, and scores"
        params:
          $ref: '#/components/schemas/QueryParams'

    QueryParams:
      type: object
      description: Optional configuration parameters for the query.
      properties:
        mode:
          type: string
          enum: [fast, standard]
          description: Extraction speed mode. Fast is quicker but may miss dynamic content.
          example: standard
        wait_for:
          type: integer
          description: Milliseconds to wait for dynamic content to load before extraction.
          example: 2000
        scroll_to_bottom:
          type: boolean
          description: If true, scroll to the bottom of the page before extraction.
          example: false
        include_screenshot:
          type: boolean
          description: If true, include a base64-encoded screenshot in the response.
          example: false
        browser_profile:
          type: string
          enum: [light, stealth]
          description: Browser profile to use for the request.
          example: light

    CreateSessionRequest:
      type: object
      description: Request body for creating a remote browser session.
      properties:
        browser_ua_preset:
          type: string
          enum: [windows, macos, linux]
          description: Operating system to simulate in the browser user agent.
          example: macos
        browser_profile:
          type: string
          enum: [light, stealth]
          description: Browser behavior profile.
          example: stealth
        inactivity_timeout_seconds:
          type: integer
          description: Seconds of inactivity before the session is automatically terminated.
          example: 300
        proxy:
          type: object
          description: Proxy configuration for the browser session.
          properties:
            type:
              type: string
              enum: [tetra, custom]
              example: tetra

    CreateSessionResponse:
      type: object
      description: Response containing the remote browser session details.
      properties:
        session_id:
          type: string
          description: Unique identifier for the created session.
          example: "sess-a1b2c3d4e5f6"
        cdp_url:
          type: string
          format: uri
          description: WebSocket URL for Chrome DevTools Protocol access.
          example: "wss://tetra.agentql.com/sessions/sess-a1b2c3d4/cdp"
        expires_at:
          type: string
          format: date-time
          description: ISO 8601 timestamp when the session will expire.
          example: "2026-04-19T15:30:00Z"

    QueryDocumentRequest:
      type: object
      description: Request body for querying data from a document.
      properties:
        url:
          type: string
          format: uri
          description: URL of the document to query (PDF or image).
          example: "https://example.com/invoice.pdf"
        query:
          type: string
          description: AgentQL query string for structured data extraction.
          example: "{ invoice { total_amount vendor_name date } }"
        prompt:
          type: string
          description: Natural language description of the data to extract.
          example: "Extract invoice totals and vendor information"

    QueryDataResponse:
      type: object
      description: Response containing extracted structured data.
      properties:
        data:
          type: object
          description: Extracted data matching the structure defined in the query.
        metadata:
          $ref: '#/components/schemas/ResponseMetadata'

    ResponseMetadata:
      type: object
      description: Metadata about the API response.
      properties:
        request_id:
          type: string
          description: Unique identifier for this API request.
          example: "req-a1b2c3d4"
        response_time_ms:
          type: integer
          description: Time taken to complete the request in milliseconds.
          example: 1847

    ErrorResponse:
      type: object
      description: Error response returned when a request fails.
      properties:
        error:
          type: string
          description: Error type or code.
          example: "invalid_query"
        message:
          type: string
          description: Human-readable error description.
          example: "The provided query is not valid AgentQL syntax."
        request_id:
          type: string
          description: Request ID for debugging.
          example: "req-a1b2c3d4"

tags:
  - name: Query Data
    description: Extract structured data from web pages using AgentQL queries
  - name: Remote Browser
    description: Manage remote Chrome browser sessions with CDP access
  - name: Query Document
    description: Extract structured data from PDF and image documents