MockMCP
Hosted MCP endpoint that returns realistic fake data for prototyping agents. Paste one URL into Claude Code, Cursor, or Claude Desktop — 12 pre-built tools covering users, products, orders, events, email, and knowledge base search. No signup, no config, no auth. Built for developers who want to prototype agent workflows before wiring up a real backend.
README
MockMCP
Hosted MCP endpoint returning realistic fake data for prototyping agents. Paste one URL, zero setup.
Live at mockmcp.io — https://mockmcp.io/mcp
Built for devs who want to prototype agent workflows without wiring up a backend, writing fixtures, or standing up a local MCP server. 12 pre-built tools covering the scenarios in most agent tutorials.
Try it
# Initialize a session and get back a session id
curl -i -X POST https://mockmcp.io/mcp \
-H "Content-Type: application/json" \
-H "Accept: application/json, text/event-stream" \
-d '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2025-06-18","capabilities":{},"clientInfo":{"name":"curl","version":"0.1"}}}'
# Use the Mcp-Session-Id header in subsequent calls:
curl -X POST https://mockmcp.io/mcp \
-H "Content-Type: application/json" \
-H "Accept: application/json, text/event-stream" \
-H "Mcp-Session-Id: <paste-id-here>" \
-d '{"jsonrpc":"2.0","id":2,"method":"tools/call","params":{"name":"list_users","arguments":{"limit":3}}}'
Connect your client
Claude Code
claude mcp add --transport http mockmcp https://mockmcp.io/mcp
Cursor — add to ~/.cursor/mcp.json:
{
"mcpServers": {
"mockmcp": {
"url": "https://mockmcp.io/mcp"
}
}
}
Claude Desktop — Settings → Connectors → Add custom connector → paste https://mockmcp.io/mcp as the URL.
No mcp-remote proxy required. All three clients speak Streamable HTTP natively.
Tools
| Tool | What it does |
|---|---|
list_users |
Paginated list of mock users |
get_user |
One mock user by id or seed |
create_user |
Create a mock user (no persistence) |
list_products |
Paginated list of mock products |
get_product |
One mock product |
list_orders |
Mock orders with optional status filter |
get_order |
One mock order |
create_order |
Create a mock order (no persistence) |
list_events |
Mock analytics events |
create_event |
Record a mock event |
send_email |
Returns fake success, nothing delivered |
search_knowledge_base |
Mock KB search with ranked results |
All data is seeded from your inputs — same input, same output.
Limits
Free tier, per hashed IP + User-Agent fingerprint:
- 30 requests per minute
- 500 requests per day
Hit the wall? Drop your email at mockmcp.io/waitlist for paid-tier access when it launches.
429 responses carry a structured JSON-RPC error with upgrade_url and retry_after_seconds. Every successful response carries X-RateLimit-Limit, X-RateLimit-Remaining, X-RateLimit-Reset, and X-RateLimit-Scope headers.
Run locally
npm install
npm run dev
Server listens on http://localhost:3000. MCP endpoint at /mcp. Without Upstash env vars, rate limiting falls back to an in-memory store (per-process, fine for dev).
Stack
@modelcontextprotocol/sdk— MCP server + Streamable HTTP transport- Raw Node
http— no framework overhead, clean shutdown @faker-js/faker— deterministic fake data via seeds@upstash/ratelimit+@upstash/redis— sliding-window rate limitingzod— tool input validation
Deploy (for forks)
Targets Node 20+. Build + start are pinned via nixpacks.toml; health checks + restart policy via railway.json.
- Fork this repo, push to GitHub.
- Railway → New Project → Deploy from GitHub → pick your fork.
- Add env vars in Railway → Variables:
| Var | Notes |
|---|---|
UPSTASH_REDIS_REST_URL |
From upstash.com free tier |
UPSTASH_REDIS_REST_TOKEN |
Paired with the URL above |
PORT |
Railway sets this automatically |
- Railway service → Settings → Networking → Custom Domain → paste your domain → follow the DNS instructions.
Health check
GET /health returns {"ok":true}. Railway pings this during deploy.
Waitlist signups
POST /api/waitlist logs a structured JSON line per signup:
{"event":"waitlist_signup","email":"alice@example.com","at":"2026-04-24T19:33:06.396Z"}
Grep them with railway logs | grep waitlist_signup. Storage-free by design — swap in Resend, a Tally forward, or a DB when the paid tier is live.
License
MIT — see LICENSE.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。