Weaviate MCP Server
MCP Server Template for Cursor IDE
一个用于使用模型上下文协议(MCP)为 Cursor IDE 创建自定义工具的简单模板。从该模板创建你自己的仓库,修改工具,并将其连接到你的 Cursor IDE。

快速开始
点击“部署到 Heroku”按钮
部署后,配置 Cursor:
打开 Cursor 设置 → 功能
添加新的 MCP 服务器
使用你的 Heroku URL,路径为
/sse(例如https://<your-app-name>.herokuapp.com/sse)
在 Cursor 中测试你的代理的心情:
询问你的代理“请询问我们服务器的心情并告诉我它怎么样。”
服务器将回复一条愉快的消息和一个心形 ❤️
Related MCP server: pure.md MCP server
其他设置方法
你可以通过三种方式运行服务器:使用 Docker、传统的 Python 设置,或直接在 Cursor IDE 中运行。
Docker 设置
项目包含 Docker 支持,便于部署:
初始设置:
# Clone the repository
git clone https://github.com/kirill-markin/weaviate-mcp-server.git
cd weaviate-mcp-server
# Create environment file
cp .env.example .env使用 Docker Compose 构建并运行:
# Build and start the server
docker compose up --build -d
# View logs
docker compose logs -f
# Check server status
docker compose ps
# Stop the server
docker compose down服务器将在以下地址可用:
SSE 端点:http://localhost:8000/sse
快速测试:
# Test the server endpoint
curl -i http://localhost:8000/sse连接到 Cursor IDE:
打开 Cursor 设置 → 功能
添加新的 MCP 服务器
类型:选择“sse”
URL:输入
http://localhost:8000/sse
传统设置
首先,安装 uv 包管理器:
# Install uv on macOS
brew install uv
# Or install via pip (any OS)
pip install uv使用 stdio(默认)或 SSE 传输启动服务器:
# Install the package with development dependencies
uv pip install -e ".[dev]"
# Using stdio transport (default)
uv run mcp-simple-tool
# Using SSE transport on custom port
uv run mcp-simple-tool --transport sse --port 8000
# Run tests
uv run pytest -v安装后,你可以直接将服务器连接到 Cursor IDE:
在 Cursor 中右键单击
cursor-run-mcp-server.sh文件选择“复制路径”以复制绝对路径
打开 Cursor 设置(齿轮图标)
导航到“功能”选项卡
向下滚动到“MCP 服务器”
点击“添加新的 MCP 服务器”
填写表单:
名称:选择任意名称(例如“my-mcp-server-1”)
类型:选择“stdio”(不是“sse”,因为我们在本地运行服务器)
命令:粘贴你之前复制的
cursor-run-mcp-server.sh的绝对路径。例如:/Users/yourname/weaviate-mcp-server/cursor-run-mcp-server.sh
环境变量
可用的环境变量(可以在 .env 中设置):
MCP_SERVER_PORT(默认值:8000)- 服务器运行的端口MCP_SERVER_HOST(默认值:0.0.0.0)- 服务器绑定的主机DEBUG(默认值:false)- 启用调试模式MCP_USER_AGENT- 用于网站抓取的自定义用户代理
其他选项
通过 Smithery 安装
要通过 Smithery 自动为 Claude Desktop 安装 MCP Server Template for Cursor IDE:
npx -y @smithery/cli install @kirill-markin/example-mcp-server --client claudeGlama 服务器审查
Available Tools
2 toolsTool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 2 tools
The two tools have completely distinct purposes: mcp_fetch is for fetching website content, while mood is a whimsical interaction unrelated to data operations. There is no overlap or ambiguity between them.
The naming is inconsistent: mcp_fetch follows a verb_noun pattern, but mood is a noun with no action verb. This mixing of conventions reduces predictability and clarity in the tool set.
With only 2 tools, the server feels severely under-scoped for a Weaviate MCP Server, which typically involves vector database operations like querying, indexing, or managing data. The tools provided do not align with the expected domain.
The tool surface is extremely incomplete for a Weaviate server. There are no tools for core vector database functions such as searching, adding data, or managing schemas, leaving significant gaps that would cause agent failures in this domain.
Maintenance
Related MCP Connectors
MCP server (stdio): fetch web pages as clean readable markdown via the AgentForge API
Turn any public website into an MCP server for agents to search, read and navigate.
Hosted Web Crawler MCP server for Claude, ChatGPT and Cursor, with each action switched per account.
Docs: https://docs.keenable.ai/mcp-server Keenable is a free, remote MCP server that gives agents access to the web index. Search the web with ranked results and date/site filters, then fetch any indexed page as clean markdown. Works out of the box with no account or API key.
Related MCP Servers
- AlicenseAqualityDmaintenanceA powerful MCP server for fetching and transforming web content into various formats (HTML, JSON, Markdown, Plain Text) with ease.49,329 npm42MIT

pure.md MCP serverofficial
AlicenseDqualityDmaintenanceAn MCP server that enables AI clients like Cursor, Windsurf, and Claude Desktop to access web content in markdown format, providing web unblocking and searching capabilities.266 npm60MIT- FlicenseNot gradedqualityFmaintenanceA lightweight MCP server implementation that provides a simple tool for fetching website content, demonstrating how tools can be exposed and interacted with using the MCP protocol.-
- AlicenseAqualityDmaintenanceAn MCP server that fetches web pages and extracts clean, AI-friendly Markdown content using Mozilla Readability. It provides secure web access for LLMs with built-in SSRF protection and automated content cleaning for improved context retrieval and summarization.180 npmMIT