{"title":"Browser Proxy","slug":"browser","category":"api","summary":"Browser automation service — separate Docker container on port 13000 for headless browser control.","audience":["human","llm"],"tags":["api","browser","automation","playwright"],"difficulty":"advanced","updated":"2026-06-27","word_count":218,"read_minutes":1,"llm_context":"Browser Proxy is a SEPARATE Docker service (port 13000), NOT a Synapse endpoint.\nSynapse endpoints do NOT include /browser/* paths.\nFor browser automation, use the MCP browser_new tool OR connect to the browser-proxy service directly.\nSynapse MCP exposes browser tools (browser_new, browser_navigate, browser_click, etc.)\n","lang":"en","translated":true,"requested_lang":"en","content_markdown":"\n# Browser Proxy\n\nThe Browser Proxy is a **separate Docker service** that provides headless\nbrowser automation via Playwright. It is NOT part of the Synapse API directly —\nSynapse endpoints do not include `/browser/*` paths.\n\n## Architecture\n\n```\n┌──────────────┐ ┌──────────────────┐\n│ Synapse API │ │ Browser Proxy │\n│ port 12800 │ │ port 13000 │\n│ │ │ (Playwright) │\n└──────────────┘ └──────────────────┘\n ▲ ▲\n │ │\n └──────────┌─────────────┐──────┘\n │ MCP Server │\n │ port 13100 │\n └─────────────┘\n```\n\n## Access Methods\n\n### Method 1: Via Synapse MCP Server (recommended)\n\nThe Synapse MCP server exposes browser tools as MCP tools. Use this for\nLLM-driven browser automation:\n\n```json\n// Claude Desktop config\n{\n \"mcpServers\": {\n \"synapse\": {\n \"command\": \"npx\",\n \"args\": [\"-y\", \"synapse-mcp-api@latest\"],\n \"env\": {\n \"SYNAPSE_MIND_KEY\": \"mk_...\",\n \"SYNAPSE_URL\": \"https://synapse.schaefer.zone\"\n }\n }\n }\n}\n```\n\nAvailable MCP browser tools:\n\n- `browser_new` — open a new browser tab\n- `browser_navigate` — navigate to a URL\n- `browser_click` — click an element\n- `browser_type` — type text into a field\n- `browser_screenshot` — capture a screenshot\n- `browser_close` — close the tab\n- (and more — see [MCP integration](/docs/mcp/what-is-mcp))\n\n### Method 2: Direct Browser Proxy Access\n\nFor non-MCP integrations, connect directly to the browser-proxy service:\n\n```bash\n# Health check\ncurl http://browser-proxy.schaefer.zone:13000/health\n\n# Open a page\ncurl -X POST http://browser-proxy.schaefer.zone:13000/navigate \\\n -H \"Content-Type: application/json\" \\\n -d '{\"url\": \"https://example.com\"}'\n```\n\n## Common Use Cases\n\n### Web scraping\n\n```python\n# Via MCP (Claude Desktop)\n\"Use browser_navigate to open https://news.ycombinator.com, \n then extract the top 10 story titles.\"\n```\n\n### Form automation\n\n```python\n# Via MCP\n\"Use browser_navigate to open the login page,\n browser_type to fill username and password,\n browser_click to submit the form.\"\n```\n\n### Screenshot capture\n\n```python\n# Via MCP\n\"Take a screenshot of https://example.com and save it.\"\n```\n\n## Related Services\n\n| Service | Port | Purpose |\n|---------|------|---------|\n| Synapse API | 12800 | Memory, chat, tasks |\n| Synapse MCP | 13100 | MCP server (79 tools) |\n| Browser Proxy | 13000 | Headless browser automation |\n| SSH Proxy | 12900 | SSH access to remote machines |\n\n## Next Steps\n\n- [MCP Integration](/docs/mcp/what-is-mcp) — how to use browser tools via MCP\n- [Computer Control API](/docs/api/computers) — for GUI automation on registered machines\n","content_html":"

Browser Proxy

\n

The Browser Proxy is a separate Docker service that provides headless\nbrowser automation via Playwright. It is NOT part of the Synapse API directly —\nSynapse endpoints do not include /browser/* paths.

\n

Architecture

\n
┌──────────────┐                ┌──────────────────┐\n│  Synapse API │                │  Browser Proxy   │\n│  port 12800  │                │  port 13000      │\n│              │                │  (Playwright)    │\n└──────────────┘                └──────────────────┘\n       ▲                                ▲\n       │                                │\n       └──────────┌─────────────┐──────┘\n                  │  MCP Server │\n                  │  port 13100 │\n                  └─────────────┘

Access Methods

\n

Method 1: Via Synapse MCP Server (recommended)

\n

The Synapse MCP server exposes browser tools as MCP tools. Use this for\nLLM-driven browser automation:

\n
// Claude Desktop config\n{\n  "mcpServers": {\n    "synapse": {\n      "command": "npx",\n      "args": ["-y", "synapse-mcp-api@latest"],\n      "env": {\n        "SYNAPSE_MIND_KEY": "mk_...",\n        "SYNAPSE_URL": "https://synapse.schaefer.zone"\n      }\n    }\n  }\n}

Available MCP browser tools:

\n\n

Method 2: Direct Browser Proxy Access

\n

For non-MCP integrations, connect directly to the browser-proxy service:

\n
# Health check\ncurl http://browser-proxy.schaefer.zone:13000/health\n\n# Open a page\ncurl -X POST http://browser-proxy.schaefer.zone:13000/navigate \\\n  -H "Content-Type: application/json" \\\n  -d '{"url": "https://example.com"}'

Common Use Cases

\n

Web scraping

\n
# Via MCP (Claude Desktop)\n"Use browser_navigate to open https://news.ycombinator.com, \n then extract the top 10 story titles."

Form automation

\n
# Via MCP\n"Use browser_navigate to open the login page,\n browser_type to fill username and password,\n browser_click to submit the form."

Screenshot capture

\n
# Via MCP\n"Take a screenshot of https://example.com and save it."

Related Services

\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
ServicePortPurpose
Synapse API12800Memory, chat, tasks
Synapse MCP13100MCP server (79 tools)
Browser Proxy13000Headless browser automation
SSH Proxy12900SSH access to remote machines
\n

Next Steps

\n\n","urls":{"html":"/docs/api/browser","text":"/docs/api/browser?format=text","json":"/docs/api/browser?format=json","llm":"/docs/api/browser?format=llm"},"translations_available":["en","zh","hi","es","fr","ar","pt","ru","ja","de","it","ko","nl","pl","tr","sv","vi","th","id","uk"]}