meloqa-mcp

meloqa-mcp

Exposes every operation from the meloQA v1 API as MCP tools, enabling LLM clients to read and manage meloQA projects, test cases, cycles, executions, bugs, links, and reference data.

Category
访问服务器

README

meloqa-mcp

Model Context Protocol (MCP) server for the meloQA public v1 API.

Exposes every operation from the v1 OpenAPI spec as an MCP tool, so LLM clients (Claude Desktop, Claude Code, Cursor, etc.) can read and manage meloQA projects, test cases, cycles, executions, bugs, links, and reference data.

Status: pre-release. Defaults to the public endpoint https://api.meloqa.com; override MELOQA_BASE_URL to point at another instance (e.g. a local http://localhost:3000).

Requirements

  • Node.js 18+ (uses native fetch).
  • A meloQA API token (see User Menu → API Tokens in the meloQA app).

Install

The package is published on npm as @meloqa/mcp-server.

You don't need to install it manually — the MCP client config below runs it on demand with npx, which fetches the latest version automatically:

npx -y @meloqa/mcp-server

To pin it globally instead (exposes the meloqa-mcp binary):

npm install -g @meloqa/mcp-server

Configure in your MCP client

Claude Desktop / Claude Code

Add an entry to your MCP config (~/Library/Application Support/Claude/claude_desktop_config.json on macOS for Claude Desktop, or .claude.json for Claude Code):

{
  "mcpServers": {
    "meloqa": {
      "command": "npx",
      "args": ["-y", "@meloqa/mcp-server"],
      "env": {
        "MELOQA_API_TOKEN": "your-token-here",
        "MELOQA_BASE_URL": "https://api.meloqa.com"
      }
    }
  }
}

Restart the client. The meloqa server should appear with ~80 tools (one per v1 endpoint).

Environment variables

Variable Required Default Description
MELOQA_API_TOKEN yes — API token sent as Authorization: Bearer <token>.
MELOQA_BASE_URL no https://api.meloqa.com Base URL of the meloQA instance. Override to point at another deployment (e.g. http://localhost:3000).
MELOQA_RATE_LIMIT no 30 Client-side limit (requests/minute) used to throttle outgoing calls. Match the server's limit.

Token lifecycle

The token sits in plaintext inside your MCP client config (e.g. ~/.claude.json). The file is chmod 600 by default, but the token still lives on disk — treat it as a credential.

To rotate (recommended periodically, and immediately if you suspect exposure):

  1. Open the meloQA UI → User Menu → API Tokens.
  2. Click Revoke on the current token. It stops working server-side within seconds.
  3. Click Generate new token and copy the new value.
  4. Replace MELOQA_API_TOKEN in your MCP client config with the new value.
  5. Restart the MCP client (Claude Desktop / Claude Code / etc.) so it re-spawns the server with the updated env.

Per-machine tokens: generate a separate token for each machine/environment so revoking one doesn't break the others.

Do not commit your MCP client config (~/.claude.json, .mcp.json, claude_desktop_config.json) to a public repo — it contains the token.

Tools

Tool names follow <resource>_<verb>, derived from the OpenAPI spec:

  • <resource>_list, <resource>_get, <resource>_create, <resource>_update, <resource>_delete for CRUD endpoints.
  • Action endpoints keep the action as the suffix: executions_run, executions_pause, executions_finish, test_cases_unarchive, links_batch_create.
  • Sub-resource listings use the parent resource as the prefix: projects_members_list, projects_execution_statuses_list, etc.

Each tool's description includes the OpenAPI summary, the tag, and any validation rules documented on the endpoint (e.g. max lengths, required referenced IDs).

Behavior notes

  • Auth header: the token is sent as Authorization: Bearer <token> (RFC 6750).
  • Rate limit: the client uses an in-process sliding-window limiter (default 30 req/min). It blocks outgoing calls until a slot is available. If you hit the server-side 429 anyway, the tool call surfaces a clear error.
  • Errors: non-2xx responses (400/401/403/404/etc.) are returned as MCP tool errors with the response body included.

Development

npm install
npm run sync-spec       # refresh spec/openapi.json from MELOQA_SPEC_URL (default http://localhost:3000/v1/docs.json)
npm run gen             # regenerate src/tools.generated.ts from the spec
npm run build           # gen + tsc → dist/
npm run dev             # run from source via tsx

To test the generated server against a running meloQA instance:

MELOQA_API_TOKEN=... MELOQA_BASE_URL=https://api.meloqa.com npm run dev

Then pipe JSON-RPC messages on stdin or wire it up via an MCP client.

License

MIT

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选