Notte Scraping API

openapi: 3.0.1
info:
  title: Notte Agents Scraping API
  description: Notte is a REST API for web browser and agent infrastructure for AI. It creates cloud browser sessions, runs autonomous web agents from natural-language tasks, exposes a perception layer to observe / step / scrape live pages, performs one-shot scraping and AI web search, and manages personas, vaults, profiles, secrets, file storage, and serverless functions. The core framework is open source (SSPL-1.0). This document models the primary documented surface; see https://api.notte.cc/openapi.json for the full machine-generated specification.
  termsOfService: https://notte.cc/terms
  contact:
    name: Notte
    url: https://notte.cc
  license:
    name: Server Side Public License v1 (SSPL-1.0)
    url: https://www.mongodb.com/licensing/server-side-public-license
  version: 1.4.40
servers:
- url: https://api.notte.cc
security:
- BearerAuth: []
tags:
- name: Scraping
  description: One-shot scraping of a URL or raw HTML, plus AI web search.
paths:
  /scrape:
    post:
      operationId: scrapeWebpage
      tags:
      - Scraping
      summary: Scrape Webpage
      description: One-shot scrape of a URL. Returns main content, links, images, or AI-extracted structured data against an optional response format.
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/GlobalScrapeRequest'
      responses:
        '200':
          description: The scraped content.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ScrapeResponse'
  /scrape_from_html:
    post:
      operationId: scrapeFromHtml
      tags:
      - Scraping
      summary: Scrape From HTML
      description: Scrapes structured content from a supplied raw HTML payload.
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/ScrapeFromHtmlRequest'
      responses:
        '200':
          description: The scraped content.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ScrapeResponse'
  /search:
    post:
      operationId: searchWeb
      tags:
      - Scraping
      summary: Search Web
      description: Runs an AI web search and returns results.
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/SearchRequest'
      responses:
        '200':
          description: The search results.
          content:
            application/json:
              schema:
                type: object
                properties:
                  results:
                    type: array
                    items:
                      type: object
components:
  schemas:
    GlobalScrapeRequest:
      allOf:
      - $ref: '#/components/schemas/ScrapeRequest'
      - type: object
        required:
        - url
        properties:
          url:
            type: string
            description: The URL to scrape.
          headless:
            type: boolean
            default: true
          solve_captchas:
            type: boolean
            default: false
          proxies:
            oneOf:
            - type: boolean
            - type: array
              items:
                type: object
          browser_type:
            type: string
            default: chromium
    ScrapeResponse:
      type: object
      properties:
        success:
          type: boolean
        content:
          type: string
          nullable: true
          description: The extracted main content (markdown).
        structured:
          type: object
          nullable: true
          description: AI-extracted structured data matching the response_format.
        links:
          type: array
          items:
            type: string
          nullable: true
        images:
          type: array
          items:
            type: string
          nullable: true
    ScrapeFromHtmlRequest:
      allOf:
      - $ref: '#/components/schemas/ScrapeRequest'
      - type: object
        properties:
          frames:
            type: array
            description: One or more HTML frames to scrape.
            items:
              type: object
              properties:
                html:
                  type: string
                url:
                  type: string
    SearchRequest:
      type: object
      required:
      - q
      properties:
        q:
          type: string
          description: The search query.
        depth:
          type: string
          description: Search depth.
        outputType:
          type: string
          description: Desired output type.
    ScrapeRequest:
      type: object
      properties:
        selector:
          type: string
          nullable: true
          description: Optional CSS selector to scope extraction.
        scrape_links:
          type: boolean
          default: false
        scrape_images:
          type: boolean
          default: false
        ignored_tags:
          type: array
          items:
            type: string
          nullable: true
        only_main_content:
          type: boolean
          default: true
        only_images:
          type: boolean
          default: false
        response_format:
          type: object
          nullable: true
          description: A JSON schema (e.g. derived from a Pydantic model) for AI-structured extraction.
        instructions:
          type: string
          nullable: true
          description: Natural-language extraction instructions.
        use_link_placeholders:
          type: boolean
          default: false
  securitySchemes:
    BearerAuth:
      type: http
      scheme: bearer
      description: Notte API key obtained from the Notte Console (https://console.notte.cc), passed as a Bearer token in the Authorization header. The SDK reads the NOTTE_API_KEY environment variable automatically.