OpenAI Tools MCP Server

ai.com.mcp由 mcp.com.ai 发布官方发布者v0.6.0更新于 Oct 24, 2026

Focused MCP server for OpenAI image/audio generation (v2.0.0). Wraps endpoints via HAPI CLI.

已验证Streamable HTTP可网页运行媒体与设计AI 与机器学习

概览

AI 生成的概览

远程 MCP 服务器，把 OpenAI 的图像、音频、模型与审核接口封装为工具，供助手调用。

可以做什么
根据提示词生成图像，编辑或扩展已有图像，并生成图像变体。把音频转写为原语言，或把音频翻译成英文。列出并查询可用模型，删除微调模型。用审核工具判断文本是否违反 OpenAI 内容政策。
适用场景
适合希望让助手生成或编辑图像、转写或翻译音频、检查文本是否符合 OpenAI 内容政策，或查看和管理 OpenAI 模型，而不必自己编写 API 调用的场景。
运行要求
通过 streamable HTTP 访问的远程端点。需要在 Authorization 请求头中提供 bearer 令牌，即从 OpenAI 控制台获取的 OpenAI API 密钥。需要能访问该端点及 OpenAI 的网络。
安装前请注意
Authorization 请求头携带 OpenAI API 密钥，按用量计费；图像、音频和审核调用会消耗付费额度。deleteModel 会永久删除微调模型，且需要组织内的 Owner 角色。提示词、图像和音频会发送给 OpenAI 以及远程端点的运营方。

安装

在 SourceWeft 中

  1. 打开 控制台中的 OpenAI Tools MCP Server，将其添加到工作区。
  2. 连接 Authorization token 凭据。它们会被私密存储，绝不会在公开页面上展示。
  3. 为需要使用其工具的对话启用该服务。

Web executable，通过 Streamable HTTP。 远程服务在工作区中配置后即可从网页运行时运行。

交给智能体安装

把这段提示词粘贴到 Claude Code、Codex、Cursor 等智能体，或 SourceWeft 对话中。智能体会读取该服务器的安装指南，先向你展示发布者、端点和登录方式，经你同意后再把它添加到你的客户端。密钥不会写进任何命令。

阅读 https://sourceweft.com/mcp/ai.com.mcp%2Fopenai-tools/install.md，并按其中说明把这个 MCP 服务器接入我的客户端。修改任何配置之前，先告诉我它的发布者、端点和登录方式，等我确认。不要把 API 密钥写进命令或我会提交的文件；问我密钥应从哪里获取。

查看智能体所依据的安装指南

其他 MCP 客户端

把它添加到你客户端的 mcpServers 配置中。

{
  "mcpServers": {
    "openai-tools": {
      "type": "http",
      "url": "https://openai-tools.run.mcp.com.ai/mcp",
      "headers": {
        "Authorization": "Bearer <YOUR_TOKEN>"
      }
    }
  }
}

README

HAPI MCP – Headless API for Model Context Protocol

Stop rewriting systems for AI. Instantly turn your APIs into MCP servers.

HAPI MCP lifts your OpenAPI catalog into MCP tools automatically, keeping design (OAS/Arazzo) and runtime (agents/LLMs) cleanly separated. Your existing APIs become AI-ready tools—no new business logic, no sidecar servers, no rework.

Key Features

  • No rewrites: Reuse 100% of your API logic. OpenAPI in, MCP tools out. No parallel codebase or shadow services.
  • Faster time-to-value: Turn specs into agent-ready tools in minutes. Update the spec, ship new tools instantly.
  • Reduced risk: Permissions, auth, rate limits, auditability are inherited from your APIs—governance without bolt-ons.
  • Scales with you: runMCP delivers serverless-like elasticity with long-running when needed. Cold-start fast, stay warm for throughput.
  • Deterministic orchestration: OrcA plans and executes multi-tool tasks predictably—no brittle prompt chaining.
  • Vendor-neutral: Works with any MCP client: ChatGPT, Claude, QBot, chatMCP. OAS + MCP + Arazzo stay portable.

How It Works

  1. Pick your API spec

    • OpenAPI, Swagger, REST—HAPI CLI works with any API specification format.
    • Supports OpenAPI 3.0+, REST APIs, OAuth 2.0 Dynamic Client Registration.

  2. Run a single command

    • One simple CLI command transforms your API into a usable MCP Server.
    • No complex setup, zero configuration, works anywhere, cross-platform.

  3. Use instantly

    • Your API is now ready as a tool, AI agent, or testing interface.
    • MCP server ready, AI agent compatible, testing interface, developer experience friendly.

Visual Pipeline

API Spec (swagger.json) → HAPI CLI → Usable Tool (AI-ready) ← MCP Clients
[ OpenAPI / Swagger ]       [ HAPI CLI ]        [ MCP Server ]       (Your API)         →   (Transform)   →   (Usable Tool)                                                 ↑          ↑                                                 |          |                                          [ ChatGPT ]  [ runMCP ]                                          [ Claude  ]  [ OrcA   ]                                          [ QBot    ]  [ chatMCP]                                          [ Agents  ]

Who Wins With HAPI

  • Executives: Ship AI initiatives without ballooning cost. Keep teams focused on outcomes, not rewrites and integration sprawl.
  • Architects & PMs: Design with OpenAPI/Arazzo, run with MCP. Clear contracts, policy inheritance, and versioned workflows keep risk low.
  • Engineers & Ops: Deploy once. HAPI Server + runMCP scale tools; QBot/chatMCP give fast feedback; OrcA keeps executions deterministic.

Your Stack, Already Wired

  • HAPI Server: Turns OpenAPI into MCP tools automatically—contracts stay in sync with your source of truth.
  • runMCP: Autoscaling execution and testing for MCP tools; cold-start fast, stay warm when workflows run long.
  • QBot: CLI TUI for power users to interact with MCP tools directly from terminal or scripts.
  • chatMCP: Conversational client that speaks MCP natively for support, ops, and internal assistants.
  • OrcA: Deterministic planning and orchestration for multi-tool tasks; no brittle prompt spaghetti.
  • Agents: Build agentic systems from standard APIs—no custom glue. Connect MCP clients across platforms.

Demos

Documentation

FAQ

  • Do we need to rewrite our services?
    No. HAPI MCP lifts your existing OpenAPI specs directly into MCP tools. Your auth, validation, and business rules remain unchanged.

  • Is this just another MCP server?
    It’s the Headless API model: your API becomes the runtime. HAPI Server reflects it as MCP; runMCP scales it; OrcA orchestrates it. No duplicate logic.

  • Which clients can consume the tools?
    Any MCP client: ChatGPT, Claude, QBot, chatMCP, bespoke orchestrators—vendor-neutral by design.

  • How do we keep control and compliance?
    Your API remains the single source of truth. Policies, RBAC, rate limits, and audit logs flow through automatically; no shadow logic.

  • What about security and privacy?
    Scoped credentials, per-tool permissions, and auditable calls are inherited from your API layer. HAPI adds guardrails and observability for regulated environments.

  • How fast can we get started?
    You can transform your first API into an MCP tool in less than five minutes. The HAPI CLI makes it quick and painless.

Get Started

  1. Install HAPI CLI from latest release: https://github.com/la-rebelion/hapimcp/releases

  2. Transform your OpenAPI spec into an MCP server with a single command:

    bash
    hapi serve path/to/your/swagger.json --headless

  3. Start using your MCP tools with any compatible client!

  4. Explore the documentation for advanced usage and best practices.

Contributing

Contributions are welcome! Open a discussion or submit a pull request on GitHub: https://github.com/la-rebelion/hapimcp

License

HAPI MCP is licensed under the MIT License. See LICENSE for details.

来源：README.md，提交 1873ff7

工具

9

createImage

Creates an image given a prompt.

输入参数createImageBody必填x-hapi-auth-state

createImageEdit

Creates an edited or extended image given an original image and a prompt.

输入参数x-hapi-auth-statecreateImageEditBody必填

createImageVariation

Creates a variation of a given image.

输入参数x-hapi-auth-statecreateImageVariationBody必填

createTranscription

Transcribes audio into the input language.

输入参数x-hapi-auth-statecreateTranscriptionBody必填

createTranslation

Translates audio into English.

输入参数x-hapi-auth-statecreateTranslationBody必填

listModels

Lists the currently available models, and provides basic information about each one such as the owner and availability.

输入参数x-hapi-auth-state

retrieveModel

Retrieves a model instance, providing basic information about the model such as the owner and permissioning.

输入参数model必填x-hapi-auth-state

deleteModel

Delete a fine-tuned model. You must have the Owner role in your organization.

输入参数model必填x-hapi-auth-state

createModeration

Classifies if text violates OpenAI's Content Policy

输入参数x-hapi-auth-statecreateModerationBody必填

版本历史

1
  1. v0.6.0最新Sep 16, 2026

评分

已满足适用检查项的 100%

该评分衡量此收录条目的完整程度，依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价，也不表示服务器安全或当前可用：可连通性是带日期的检查结果，评分也不包含安装量或用户评价。

  • 近 30 天内可连通必需25 分
  • 已声明认证方式必需10 分
  • 已列出工具必需15 分
  • 描述必需15 分
  • 图标10 分
  • 文档15 分
  • 许可证10 分