MCP · mcp://api.ujeebu.com

One MCP endpoint. Eight tools.

A native Model Context Protocol server: every endpoint exposed as a typed tool, ready to plug into Claude Desktop, Cursor, Windsurf, or any MCP-compliant agent. Bring your API key, get web access in 60 seconds.

Try in playground Read the docs

tools exposed

60s

install to first call

∞

agents supported

MCP mcp://api.ujeebu.com

# claude_desktop_config.json
{
  "mcpServers": {
    "ujeebu": {
      "command": "npx",
      "args":    ["-y", "@ujeebu/mcp"],
      "env":     { "UJEEBU_KEY": "uj_live_..." }
    }
  }
}

# In Claude:
> Find me the top 5 React component
> libraries by GitHub stars and
> summarize their differences.

🛠 ujeebu.serp({ q: "react component library" })
🛠 ujeebu.scrape(url: github.com/...)  ×5
🛠 ujeebu.extract(url: ...)              ×5

Native MCP

Real Model Context Protocol, not a wrapper, not a translation layer. Tools, resources, and prompts conform to the spec.

8 tools, typed

scrape, extract, ai_scraper, serp, markdown, chatgpt, gemini, account. Each ships with full JSON Schema so the agent picks correctly.

Per-key budgets

Cap how many credits an agent can burn per session. Hard stop at the limit. No nightmare bills from a rogue eval loop.

Citations built in

Every tool result includes the source URL, fetch timestamp, and credit cost. Your agent can ground its claims; you can audit them.

Streaming results

For long batch runs, results stream into the agent's context as they arrive, so the model can start reasoning before the full call returns.

Multi-client

Works with Claude Desktop, Cursor, VS Code, Windsurf, the OpenAI Agents SDK, Google ADK, and any other MCP-compatible host. Same install, every client.

Set up in 60 seconds

Pick your client. Paste the config. Restart.

Same MCP server, every host. Cursor and VS Code get a one-click install. Everyone else copies a config block or drops an SDK snippet into their code.

Your Ujeebu API key

Stays in your browser. Every snippet below regenerates as you type.

File location ~/Library/Application Support/Claude/claude_desktop_config.json macOS

%APPDATA%\Claude\claude_desktop_config.json Windows

Then quit and reopen Claude Desktop.

File location ~/.cursor/mcp.json global .cursor/mcp.json per-project

Install in Cursor

Or paste into the config file and reload Cursor.

File location ~/.vscode/mcp.json user .vscode/mcp.json workspace

Install in VS Code

Requires VS Code 1.95+ with MCP support enabled.

File location ~/.codeium/windsurf/mcp_config.json

Then reload Windsurf (Cmd-Shift-P → Reload Window).

Drop into your agent script pip install openai-agents

Tools auto-discovered from the MCP server: scrape, serp, extract, and more.

Works with any OpenAI-compatible model — DeepSeek, Kimi, Qwen, Groq, vLLM, Ollama. Just point an AsyncOpenAI client at the provider's base_url and pass it to OpenAIChatCompletionsModel.

Drop into your agent script pip install google-adk

Add the toolset to any ADK Agent. Works with Gemini, Vertex, or self-hosted models.

Any MCP-compatible host stdio transport

Point your host at the stdio command with UJEEBU_KEY in env.

Stuck? Read the MCP setup guide or ping support.

What people build with it

Real things real teams shipped this quarter.

Agentic web research

A junior researcher asks Claude Desktop to compare four GitHub libraries. The agent runs SERP → article → markdown → reasons over the result. 90 seconds end-to-end.

IDE-side fetches

In Cursor, ask "what does Stripe's API say about idempotency keys?" The MCP server fetches, summarizes, and pastes the source URL into your editor.

Custom agent products

Build a vertical agent (legal research, sales prospecting). The MCP server gives it instant web access; you focus on the domain logic.

Eval harnesses

Wire MCP into your eval framework. Per-key budgets stop runaway test loops; full request logs let you reproduce any agent step exactly.

Ship MCP Server tonight.

5,000 credits free. No card. Real residential proxies on the free tier.

Get API key Try in playground