mcpwrap
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@mcpwrapwrap https://petstore3.swagger.io/api/v3/openapi.json"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
mcpwrap
Live: https://mcp-forge-black.vercel.app

Turn any REST API into an MCP server in 10 seconds.
npx mcpwrap --spec https://petstore3.swagger.io/api/v3/openapi.jsonNo code generation. No config files. Just point and serve.
Why?
Other tools generate static code from your OpenAPI spec. mcpwrap wraps your API at runtime - it reads the spec, creates MCP tools dynamically, and proxies requests to your API. Update the spec, restart, done.
Related MCP server: ControlAPI-MCP
Quick Start
# From a URL
npx mcpwrap --spec https://api.example.com/openapi.json
# From a local file
npx mcpwrap --spec ./my-api.yaml
# With authentication
npx mcpwrap --spec ./api.json --auth "Bearer $MY_TOKEN"
# With API key header
npx mcpwrap --spec ./api.json --auth "X-API-Key:your-key-here"
# Filter endpoints
npx mcpwrap --spec ./api.json --include "/pets/*" --exclude "/admin/*"Claude Desktop Integration
Add to your Claude Desktop config (claude_desktop_config.json):
{
"mcpServers": {
"my-api": {
"command": "npx",
"args": ["mcpwrap", "--spec", "https://api.example.com/openapi.json"]
}
}
}How It Works
Reads your OpenAPI 3.0/3.1 spec (JSON or YAML, local or remote)
Parses every endpoint into an MCP tool (one tool per operation)
Maps parameters and request bodies to MCP tool input schemas
Proxies tool calls to your API and returns the response
Flattens large/nested responses for LLM readability
Features
Runtime wrapping - no generated code to maintain
OpenAPI 3.0 and 3.1 support
All HTTP methods - GET, POST, PUT, DELETE, PATCH
Auth passthrough - Bearer tokens, API keys, Basic auth
Path and query params - automatically mapped from tool arguments
Request body - JSON body passed as
bodyargumentResponse flattening - truncates deep nesting and long arrays
Endpoint filtering - include/exclude by path pattern
Verbose mode - logs requests to stderr for debugging
CLI Options
Option | Description | Default |
| Path or URL to OpenAPI spec (required) | - |
| Override base URL from spec | From spec |
| Auth: | None |
| Only include matching paths | All |
| Exclude matching paths | None |
|
|
|
| Port for HTTP transport |
|
| Disable response flattening | Enabled |
| Log requests to stderr | Off |
Tool Naming
Tools are named using the operationId from your spec when available, falling back to METHOD_path_segments (e.g., get_pets_petId).
Development
git clone https://github.com/m4cd4r4/mcpwrap
cd mcpwrap
npm install
npm test
npm run dev -- --spec examples/petstore.json --verboseLicense
MIT
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Latest Blog Posts
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/m4cd4r4/mcpwrap'
If you have feedback or need assistance with the MCP directory API, please join our Discord server