# I Built 3 MCP Servers for AI Agents — Here's How They Work

> Source: <https://dev.to/kenneth_doster_5d86c3cf4f/i-built-3-mcp-servers-for-ai-agents-heres-how-they-work-5dbj>
> Published: 2026-06-28 06:16:10+00:00

The Model Context Protocol (MCP) is an open standard that lets AI agents use external tools through a unified interface. Think of it as USB-C for AI — one protocol connects any AI client (Claude Desktop, Cursor, VS Code with Cline) to any tool or data source.

I built three production-ready MCP servers and published them to PyPI and GitHub. Here's what they do and how to use them.

`uvx crewai-web-search-mcp`

Two tools:

Use cases: Ask your AI about current events, research competitors, pull documentation, verify facts in real time.

```
{
  "mcpServers": {
    "web-search": {
      "command": "uvx",
      "args": ["crewai-web-search-mcp"]
    }
  }
}
```

`uvx code-review-automation`

Three tools:

Use cases: Paste a PR diff and get an instant review. Catch issues before they reach production.

`uvx document-intelligence-server`

Three tools:

Use cases: Process uploaded PDFs, extract data from scanned forms, summarize long reports.

All three servers use a shared credit system:

| Tier | Price | Credits |
|---|---|---|
| Free | $0 | 50 calls/day |
| Starter | $20 | 2,000 calls |
| Pro | $100 | 12,000 calls |

Buy credits once, use them across any server. Credits never expire.

**How it works:**

`uvx crewai-web-search-mcp`

`MCP_LICENSE_KEY`

→ all limits removedThe billing backend is open source too: [github.com/KennyWayn3/mcp-billing-api](https://github.com/KennyWayn3/mcp-billing-api)

`pip install crewai-web-search-mcp`

What MCP servers would you like to see built next? I'm curious what the community actually needs.