# Browser Proxy

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

## Architecture

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

## Access Methods

### Method 1: Via Synapse MCP Server (recommended)

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

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

Available MCP browser tools:

- `browser_new` — open a new browser tab
- `browser_navigate` — navigate to a URL
- `browser_click` — click an element
- `browser_type` — type text into a field
- `browser_screenshot` — capture a screenshot
- `browser_close` — close the tab
- (and more — see [MCP integration](/docs/mcp/what-is-mcp))

### Method 2: Direct Browser Proxy Access

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

```bash
# Health check
curl http://browser-proxy.schaefer.zone:13000/health

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

## Common Use Cases

### Web scraping

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

### Form automation

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

### Screenshot capture

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

## Related Services

| Service | Port | Purpose |
|---------|------|---------|
| Synapse API | 12800 | Memory, chat, tasks |
| Synapse MCP | 13100 | MCP server (79 tools) |
| Browser Proxy | 13000 | Headless browser automation |
| SSH Proxy | 12900 | SSH access to remote machines |

## Next Steps

- [MCP Integration](/docs/mcp/what-is-mcp) — how to use browser tools via MCP
- [Computer Control API](/docs/api/computers) — for GUI automation on registered machines