> ## Documentation Index > Fetch the complete documentation index at: https://docs.55-tech.com/llms.txt > Use this file to discover all available pages before exploring further. # Live browser session with event streaming (SSE) > Open a browser, navigate to the target URL, and stream back events in real-time via Server-Sent Events. Only captures XHR/Fetch API responses and WebSocket frames — HTML documents and scripts are filtered out. No resources are blocked by default so pages load fully. Default wait strategy is networkidle. All data is gzip-compressed. The session runs until the client disconnects (24h safety cap). Set X-Timeout to limit the duration. ## OpenAPI ````yaml /zh/scraping-api/openapi.json get /browser/stream openapi: 3.1.0 info: title: 55 Tech Scraping API description: >- Distributed fetch network. Route HTTP, browser, WebSocket, and AMQP requests through 80+ geo-distributed agents with health-aware routing and block detection. version: 2.2.0 contact: name: 55 Tech Support email: contact@55-tech.com servers: - url: https://scraping-api.55-tech.com description: Production security: - ApiKeyHeader: [] tags: - name: Fetch description: Proxy HTTP requests through the agent network. - name: Browser description: Fetch with full JavaScript rendering, cookies, and screenshots. - name: Network description: Agent registry, network status, and domain health. - name: AMQP description: AMQP/RabbitMQ consumer relay via SSE. - name: Usage description: Per-key usage metrics. - name: Debug description: Debug and diagnostic endpoints. - name: System description: Health checks and system endpoints. paths: /browser/stream: get: tags: - Browser summary: Live browser session with event streaming (SSE) description: >- Open a browser, navigate to the target URL, and stream back events in real-time via Server-Sent Events. Only captures XHR/Fetch API responses and WebSocket frames — HTML documents and scripts are filtered out. No resources are blocked by default so pages load fully. Default wait strategy is networkidle. All data is gzip-compressed. The session runs until the client disconnects (24h safety cap). Set X-Timeout to limit the duration. operationId: browser_stream parameters: - $ref: '#/components/parameters/XTargetURL' - name: X-Capture in: header description: 'Comma-separated event types to stream: network, ws, dom, console.' schema: type: string default: network,ws,console - name: X-DOM-Selector in: header description: CSS selector to watch for DOM mutations (requires dom in X-Capture). schema: type: string - name: X-Network-Filter in: header description: Regex filter for network URLs. Only matching responses are streamed. schema: type: string - name: X-WS-Filter in: header description: >- Regex filter for WebSocket URLs. Only matching connections are streamed. schema: type: string - name: X-JS-After-Load in: header description: >- JavaScript to execute after page load (e.g. click a button, trigger navigation). schema: type: string - name: X-Steps in: header description: >- JSON array of sequential browser actions (navigate, click, type, wait, etc.). Executes after page load, before streaming starts. Step results arrive as step_ok/step_error SSE events. schema: type: string - name: X-Wait-Strategy in: header description: >- When the page is ready before streaming starts: load, networkidle, or selector. schema: type: string default: networkidle - name: X-Wait-Selector in: header description: 'CSS selector to wait for (with X-Wait-Strategy: selector).' schema: type: string - name: X-Proxy in: header description: Route through a proxy (http:// or socks5://). schema: type: string - name: X-Block-Resources in: header description: >- Comma-separated resource types to block: image, font, stylesheet, media. No default blocking — pages load fully. schema: type: string - name: Cookie in: header description: Cookies to inject (name=value; name2=value2). schema: type: string - name: X-Cookies in: header description: Cookies as JSON array for full control. schema: type: string - $ref: '#/components/parameters/XGeo' - $ref: '#/components/parameters/XAgent' - $ref: '#/components/parameters/XTimeout' responses: '200': description: >- SSE event stream. Events: connected, network, ws_open, ws_message, ws_close, dom, console, error, done. content: text/event-stream: schema: type: string example: >+ event: connected data: {"type":"connected","url":"https://example.com","agent_id":"scraping-us5"} event: ws_message data: {"type":"ws_message","data":"{\"price\":1.95}","ws_id":"1234"} '401': description: Missing API key. '403': description: Invalid API key. '429': description: Rate limit exceeded. '502': description: No healthy agent available. components: parameters: XTargetURL: name: X-Target-URL in: header description: >- Target URL (no encoding needed). Recommended for URLs with query parameters. schema: type: string XGeo: name: X-Geo in: header description: >- Target countries as comma-separated ISO codes (e.g. US,DE,AT). Aliases: X-Geo-CC, X-CC, X-Country, X-Geo-Strict. schema: type: string XAgent: name: X-Agent in: header description: >- Route through specific agent(s) by slug (e.g. de1) or comma-separated list for random pick (e.g. de1,at5,us3). schema: type: string XTimeout: name: X-Timeout in: header description: 'Override request timeout in seconds (default: 30).' schema: type: string securitySchemes: ApiKeyHeader: type: apiKey in: header name: X-API-Key description: API key passed in the X-API-Key header. ````