
Postman
Postman’s remote MCP server connects AI agents, assistants, and chatbots directly to your APIs on Postman. Use natural language to prompt AI to automate work across your Postman collections, environments, workspaces, and more.
README
Postman MCP Server
This project offers the following Model Context Protocol (MCP) server options:
For more information about the available transports, see the MCP specification.
STDIO
This is a lightweight solution that's ideal for integration with editors and tools like VS Code.
For Docker set up and installation, see DOCKER.md.
VS Code integration
Note: By default, this server provides 37 tools (minimal mode). Use the
--full
flag to access all 106 tools.
Integrate your MCP server with Visual Studio Code and use it with VS Code extensions that support MCP. To do this, do the following:
-
Create a .vscode/mcp.json file in your project and enter the following:
{ "servers": { "postman-api-mcp": { "type": "stdio", "command": "npx", "args": [ "@postman/postman-mcp-server", "--full" // (optional) Use this flag to enable full mode ], "env": { "POSTMAN_API_KEY": "${input:postman-api-key}" } } }, "inputs": [ { "id": "postman-api-key", "type": "promptString", "description": "Enter your Postman API key" } ] }
-
Install an MCP-compatible VS Code extension, such as GitHub Copilot, Claude for VS Code, or other AI assistants that support MCP.
Configure the extension
Configure the extension to use the postman-api-mcp server, a local STDIO-based server that runs directly from your project files:
- Clone the postman-mcp-server repository.
- In the repository's root folder, run the
npm install
command. This installs all the required dependencies. - Replace
${workspaceFolder}
in the mcp.json file with the full path to the Postman MCP repository. - When prompted, enter your Postman API key.
Claude integration
To integrate the MCP server with Claude, check the latest Postman MCP server release and get the .dxt
file. For more information, see Anthropic's Claude Desktop Extensions documentation.
Streamable HTTP
The streamable HTTP version is available at https://mcp.postman.com
. It supports two tool configurations to better serve different use cases:
- Minimal — Only includes essential tools for basic Postman operations, available at
https://mcp.postman.com/minimal
. This offers faster performance and simplifies use for those who only need basic Postman operations. - Full — Includes all available Postman API tools (100+ tools), available at
https://mcp.postman.com/mcp
.
Cursor integration
To integrate the MCP server with Cursor, click the following button:
Ensure the Authorization header uses the Bearer <YOUR_API_KEY> format.
VS Code integration
By default, the server provides 37 tools. Use Full (
https://mcp.postman.com/mcp
) mode to access all 106 tools.
To install in VS Code, you can use the Postman VS Code Extension. Or you can add the following to the .vscode/mcp.json file:
{
"servers": {
"postman-api-http-server": {
"type": "http",
"url": "https://mcp.postman.com/{minimal | mcp}", // choose "minimal" or "mcp"
"headers": {
"Authorization": "Bearer ${input:postman-api-key}"
}
}
},
"inputs": [
{
"id": "postman-api-key",
"type": "promptString",
"description": "Enter your Postman API key"
}
]
}
When prompted, enter your Postman API key. Afterwards, the agent performs calls to the Postman cloud MCP server at https://mcp.postman.com
.
Migration from v1.x to v2.x
- Tool naming changes - All tool names changed from kebab-case to camelCase. For example:
create-collection
→createCollection
get-workspaces
→getWorkspaces
delete-environment
→deleteEnvironment
- Tool availability changes
- The default (Minimal) behavior provides only 37 essential tools.
- The
--full
flag provides access to all 106 tools.
Questions and support
- See the Postman Agent Generator page for updates and new capabilities.
- See Add your MCP requests to your collections to learn how to use Postman to perform MCP requests.
- Visit the Postman Community to share what you've built, ask questions, and get help.
- You can connect to both HTTP and STDIO servers and test them using the Postman MCP Server collection.
推荐服务器

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。
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。
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。
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。

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。
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。
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。

e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
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。