Residential proxy MCP for AI agents — geo-targeting, sticky sessions, anti-bot bypass.
Connect Proxy4agents to Claude, Cursor or any other MCP client and it stops being a tab you switch to. Residential proxy MCP for AI agents — geo-targeting, sticky sessions, anti-bot bypass. The proxy4agents mcp server is what makes that connection.
The agent-first residential proxy MCP — works with any provider.
The server ships on npm as novada-proxy-mcp, so your MCP client can launch it on demand — there is no separate build step. Add the server block to your client's configuration, restart it, and the tools register themselves.
The toolset is worth reading before you wire it up, because it tells you what the integration is really for:
novada_proxy_fetch — Fetch any URL through residential proxynovada_proxy_batch_fetch — Fetch 2–20 URLs concurrently (up to 5x parallel)novada_proxy_extract — Extract structured fields — heuristic mode (fields) or LLM mode (schema)novada_proxy_map — Crawl a URL and return all internal links as JSON arraynovada_proxy_crawl — Recursively crawl a site (BFS, depth 1-5) with URL discoverynovada_proxy_session — Sticky session — same IP across every callnovada_proxy_search — Google search -> structured JSON (title, url, snippet)novada_proxy_render — Render JS-heavy pages with real Chromium [BETA]novada_proxy_research — One-shot deep research — search + fetch + synthesizenovada_proxy_status — Check proxy network health + versionConfiguration is passed through the environment: NOVADA_PROXY_USER, NOVADA_PROXY_PASS, NOVADA_API_KEY, BRIGHTDATA_HOST, SMARTPROXY_HOST, OXYLABS_HOST, PROXY_URL. Treat anything key-shaped as a real credential — scope it to the minimum the server needs, and rotate it if it ever lands in a shared config.
Plenty of search and retrieval servers cover similar ground. The differences that matter in practice are scope of access and how much setup stands between you and a working tool call. Proxy4agents's toolset — novada_proxy_fetch, novada_proxy_batch_fetch, novada_proxy_extract and 7 more — is a fair guide to whether it matches your workflow. It is maintained by NovadaLabs; worth a glance at recent repository activity before you build anything load-bearing on it.
We check each listing at SyncDev against the project's documentation before it goes live — if something here drifts out of date, it is a bug worth reporting.
| Tool | What it does |
|---|---|
| novada_proxy_fetch | Fetch any URL through residential proxy |
| novada_proxy_batch_fetch | Fetch 2–20 URLs concurrently (up to 5x parallel) |
| novada_proxy_extract | Extract structured fields — heuristic mode (fields) or LLM mode (schema) |
| novada_proxy_map | Crawl a URL and return all internal links as JSON array |
| novada_proxy_crawl | Recursively crawl a site (BFS, depth 1-5) with URL discovery |
| novada_proxy_session | Sticky session — same IP across every call |
| novada_proxy_search | Google search -> structured JSON (title, url, snippet) |
| novada_proxy_render | Render JS-heavy pages with real Chromium [BETA] |
| novada_proxy_research | One-shot deep research — search + fetch + synthesize |
| novada_proxy_status | Check proxy network health + version |
{
"mcpServers": {
"proxy4agents": {
"command": "npx",
"args": ["-y", "novada-proxy-mcp"],
"env": {
"NOVADA_PROXY_USER": "your-value",
"NOVADA_PROXY_PASS": "your-value",
"NOVADA_API_KEY": "your-value",
"BRIGHTDATA_HOST": "your-value",
"SMARTPROXY_HOST": "your-value",
"OXYLABS_HOST": "your-value",
"PROXY_URL": "your-value"
}
}
}
}Add to claude_desktop_config.json, then restart Claude Desktop.
| Variable | Description | Required |
|---|---|---|
| NOVADA_PROXY_USER | Configuration value read at startup. | Optional |
| NOVADA_PROXY_PASS | Configuration value read at startup. | Optional |
| NOVADA_API_KEY | Credential the server authenticates with. | Yes |
| BRIGHTDATA_HOST | Endpoint or connection string the server talks to. | Optional |
| SMARTPROXY_HOST | Endpoint or connection string the server talks to. | Optional |
| OXYLABS_HOST | Endpoint or connection string the server talks to. | Optional |
| PROXY_URL | Endpoint or connection string the server talks to. | Yes |
The simplest web tool that matters — fetch any URL and get model-ready markdown back.
Industrial-strength web extraction — render, scrape, crawl and search entire sites into clean markdown.
Puppeteer-powered browser control that drives pages from the accessibility tree instead of pixels.
Search built for AI, not humans — semantic web search that returns model-ready content, plus code context.
Answers, not links — delegate questions to Perplexity's search-grounded models and get cited responses back.
Put 6,000+ pre-built scrapers at your assistant's fingertips through one MCP endpoint.