How do I connect the server to Cursor IDE?
Open Cursor Settings → Features, add a new MCP server, choose “sse” (for remote/Docker) or “stdio” (for local), and enter the appropriate URL or command path.
@netr
基本情報
概要はまだありません
以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。
{
"mcpServers": {
"mcp-server-learn": {
"command": "docker",
"args": [
"compose",
"up",
"--build",
"-d"
]
}
}
}ツール一覧はまだありません
この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。
ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。
A simple template for creating custom tools for Cursor IDE using Model Context Protocol (MCP). It allows developers to create their own repository from this template, modify the tools, and connect them to Cursor IDE.
Deploy the server to Heroku with one click, or run locally via Docker or traditional Python setup. Configure Cursor IDE by adding a new MCP server with either an SSE endpoint (Heroku or Docker) or a stdio command (local setup).
Open Cursor Settings → Features, add a new MCP server, choose “sse” (for remote/Docker) or “stdio” (for local), and enter the appropriate URL or command path.
The server supports two transports: SSE (Server-Sent Events) for remote or Docker setups, and stdio for local execution.
Set `MCP_SERVER_PORT` (default 8000), `MCP_SERVER_HOST` (default 0.0.0.0), `DEBUG` (default false), and `MCP_USER_AGENT` in a `.env` file.
The README only provides a Heroku deployment button; other platforms are not mentioned.
The server uses Python with `uv` package manager and can be installed with `uv pip install -e ".[dev]"`. Docker is an alternative deployment method.

Automated codebase documentation for GitHub: searchable docs, MCP context for AI agents, doc drift detection, and Friday Cleanup PRs.

Manage your Skycloak-hosted Keycloak: clusters, realms, apps, SSO and users, from any MCP client.

Routebase keeps your API's design, docs, contract tests, mocks and monitoring in one living OpenAPI spec so they can't drift apart. Agents work on that same source of truth, under your team's permissions.

Agents make claims. Reelier writes receipts — record an agent's tool-call workflow once, replay it deterministically at 0 tokens, and diff runs to catch drift.

Verified distributed compute for humans, applications, and agents. Submit Blender, Octane, and video workloads, route them across qualified GPUs, track asynchronous execution, download finished artifacts, and verify FARP
コメント