How do I customize authentication?
Override the `get-axios.ts` file with your own authentication logic. By default, it uses a Bearer token from the `API_KEY` environment variable. With Docker, mount a custom version of `get-axios.ts` as a volume.
@nexpando-com
基本信息
Generate a MCP server based on OpenAPI specifications
使用下面的配置,将此服务器添加到你的 MCP 客户端。
{
"mcpServers": {
"open-api-mcp": {
"command": "bun",
"args": [
"dev"
]
}
}
}工具列表暂未提供
当前收录暂时没有可展示的工具列表,不代表服务没有工具。请查阅项目文档了解支持的能力。
工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。
open-api-mcp is a command-line tool that generates an MCP (Model Context Protocol) server from OpenAPI specifications. It simplifies creating API clients by leveraging OpenAPI schemas and generating strongly-typed clients using Zod.
Clone the repository, install dependencies with bun install, prepare an OpenAPI specification (JSON or YAML), copy .env.template to .env, then start the server with bun dev or ./cli.sh open-api.json. For Docker, add the spec to a specs directory and run docker-compose up with appropriate environment variables.
Override the `get-axios.ts` file with your own authentication logic. By default, it uses a Bearer token from the `API_KEY` environment variable. With Docker, mount a custom version of `get-axios.ts` as a volume.
Supported transport types are `httpStream`, `sse`, and `stdio`, configurable via the `MCP_TRANSPORT_TYPE` environment variable.
Create a `specs` directory with your OpenAPI file, define a `docker-compose.yml` with the `nexpando/open-api-mcp` image, mount the specs directory, and set environment variables (`OPEN_API_FILE`, `API_URL`, optionally `API_KEY`, `MCP_NAME`, `MCP_VERSION`, `MCP_TRANSPORT_TYPE`). Then run `docker-compose up`.
The project uses the Bun runtime (install with `bun install`). A Docker image is also available for containerized deployment.
Place it in the `specs` directory (e.g., `specs/open-api.json` or `specs/open-api.yml`) and set the `OPEN_API_FILE` environment variable to point to it.

Modelglass turns "which model should I use?" into a lookup instead of a guess. Seven tools give your agent live pricing across image, language, video, and audio models. Sourced and versioned, not scraped once and forgott

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

ArtiBot Cloud MCP (Zhijiangyun 智匠云) — cloud MCP server for robot manipulation data & GPU workloads. 15 tools: robot fleet management, data collection & video recording, post-processing pipelines, and benchmark evaluation

Local-first credential broker for MCP. AI coding agents (Claude Code, Cursor, Codex…) call APIs, SSH and databases through scoped presets; secrets stay in your OS keychain and never enter the agent's context.
评论