# Ai Eval

> Ai Eval is an MCP server for ai models and llm tooling. Remote MCP server and REST API named ai-eval, running on Cloudflare Workers. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Platforms:** API
- **Repository:** https://github.com/lazymac2x/ai-eval-api
- **MCP Registry name:** io.github.lazymac2x/ai-eval
- **Categories:** [AI models and LLM tooling](https://ai-powered.com/categories/ai-and-llm)
- **Listing updated:** 2026-11-18

## About

Ai Eval runs on Cloudflare Workers and can be reached in two ways: as a plain REST API and as a Model Context Protocol endpoint that speaks JSON-RPC 2.0. A hosted instance is published at api.lazy-mac.com/ai-eval.

MCP clients connect over streamable HTTP to the remote URL https://api.lazy-mac.com/ai-eval/mcp, so nothing has to be installed locally. The registry entry and README give only a short description and do not list the individual tools it exposes or what it evaluates.

It suits developers who want to try a lightweight hosted endpoint from an MCP-capable client or call the same service directly over HTTP. Check the repository for current details before relying on it.

## Features

- **Remote MCP endpoint:** Exposes an MCP server over streamable HTTP at a hosted URL, so no local install is needed.
- **REST and MCP access:** Serves both a conventional REST interface and MCP using JSON-RPC 2.0.
- **Cloudflare Workers runtime:** Built to run on Cloudflare Workers, a serverless edge platform.
- **Public hosted instance:** A live deployment is available at api.lazy-mac.com/ai-eval.

## Good for

- Connect an MCP-capable AI client to a hosted remote server without local setup
- Call the same service over REST from scripts or backend code
- Use as a reference for deploying an MCP server on Cloudflare Workers

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http ai-eval https://api.lazy-mac.com/ai-eval/mcp
```

### Claude Desktop (Settings → Connectors → Add custom connector)

```sh
https://api.lazy-mac.com/ai-eval/mcp
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "ai-eval": {
      "url": "https://api.lazy-mac.com/ai-eval/mcp"
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "ai-eval": {
      "type": "http",
      "url": "https://api.lazy-mac.com/ai-eval/mcp"
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "ai-eval": {
      "serverUrl": "https://api.lazy-mac.com/ai-eval/mcp"
    }
  }
}
```

### Grok Bot (Grok Bot desktop app → Connect apps → Add a custom connector)

```sh
https://api.lazy-mac.com/ai-eval/mcp
```

Enter a Name (for example "Ai Eval"), choose Type: Remote HTTPS, paste this as the Server URL and choose Add plugin.

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "ai-eval": {
      "type": "http",
      "url": "https://api.lazy-mac.com/ai-eval/mcp"
    }
  }
}
```

## Tools

- `score_response`: Score an AI response against a prompt using heuristic metrics (length, relevance, structure, completeness)
- `compare_responses`: Compare and rank multiple AI responses to the same prompt
- `text_metrics`: Get text quality metrics: word count, sentence count, estimated tokens, readability grade

## FAQ

### How do I connect to Ai Eval over MCP?

Point an MCP client at the streamable HTTP remote https://api.lazy-mac.com/ai-eval/mcp.

### Does it also offer a REST API?

Yes. The README says it provides both REST and MCP using JSON-RPC 2.0.

### Where does it run?

It is a Cloudflare Workers server, with a live instance at api.lazy-mac.com/ai-eval.

### How do I add Ai Eval to Claude Code?

Run this command in your terminal: claude mcp add --transport http ai-eval https://api.lazy-mac.com/ai-eval/mcp

### What are the best alternatives to Ai Eval?

The closest alternatives to Ai Eval are Was ist MCP? Demo-Server, Hello, AI and EvalGate Docs.

## Where this comes from

- Description: written from the project's README and its MCP Registry entry.
- Status and tools: checked automatically by AIPoweredBot on 19 November 2026.
- Install steps: from the official MCP Registry entry.

## Alternatives

- [Was ist MCP? Demo-Server](https://ai-powered.com/mcp/was-ist-mcp-demo-server): Was ist MCP? Demo-Server is a free MCP server for ai models and llm tooling.
- [Hello, AI](https://ai-powered.com/mcp/hello-ai): Hello, AI is an MCP server for ai models and llm tooling.
- [EvalGate Docs](https://ai-powered.com/mcp/evalgate-docs): EvalGate Docs is a freemium MCP server for ai models and llm tooling.

---
Source: https://ai-powered.com/mcp/ai-eval
