Wrap any existing Echo API into MCP tools, enabling AI agents to interact with your API through Model Context Protocol. Inspired by gin-mcp but for the Echo framework. - Zero Configuration: Works with any existing Echo API - Multiple Schema Sources: Support for Swaggo, raw OpenAPI YAML/JSON, and manual schemas - Filtering: Include/exclude endpoints with wildcard patterns - MCP Compatible: Works wi
Add this skill
npx mdskills install BrunoKrugel/echo-mcpTransforms Echo APIs into MCP-compatible tools with multiple schema sources and flexible endpoint filtering
Wrap any existing Echo API into MCP tools, enabling AI agents to interact with your API through Model Context Protocol.
Inspired by gin-mcp but for the Echo framework.
go get github.com/BrunoKrugel/echo-mcp
package main
import (
"net/http"
server "github.com/BrunoKrugel/echo-mcp"
"github.com/labstack/echo/v4"
)
func main() {
e := echo.New()
// Existing API routes
e.GET("/ping", func(c echo.Context) error {
return c.JSON(http.StatusOK, map[string]string{"message": "pong"})
})
// Add MCP support
mcp := server.New(e)
mcp.Mount("/mcp")
e.Start(":8080")
}
Now the API is accessible at http://localhost:8080/mcp
If you already use Swaggo for Swagger documentation, enable automatic schema generation:
// @Summary Get user by ID
// @Description Retrieve detailed user information
// @Tags users
// @Param id path int true "User ID" minimum(1)
// @Success 200 {object} User
// @Router /users/{id} [get]
func GetUser(c echo.Context) error {
// Your handler code
}
func main() {
e := echo.New()
e.GET("/users/:id", GetUser)
// Enable automatic swagger schema generation
mcp := server.NewWithConfig(e, &server.Config{
BaseURL: "http://localhost:8080",
EnableSwaggerSchemas: true,
})
mcp.Mount("/mcp")
e.Start(":8080")
}
If you use other OpenAPI libraries like swaggest/openapi-go, you can pass a raw YAML or JSON schema string:
import (
"github.com/swaggest/openapi-go/openapi3"
server "github.com/BrunoKrugel/echo-mcp"
)
func main() {
e := echo.New()
// ... define your routes ...
// Generate OpenAPI schema
reflector := openapi3.Reflector{}
reflector.SpecEns().WithOpenapi("3.0.3")
reflector.SpecEns().Info.WithTitle("My API").WithVersion("1.0.0")
// Add operations to reflector
// ...
// Export to YAML (or JSON)
schema, _ := reflector.Spec.MarshalYAML()
// Pass raw schema to MCP server
// ... you can also embed the schema from an exiting openapi.yaml file
mcp := server.NewWithConfig(e, &server.Config{
OpenAPISchema: string(schema),
})
mcp.Mount("/mcp")
e.Start(":8080")
}
The OpenAPISchema field accepts both YAML and JSON formatted strings. When provided, it automatically populates:
Expose only the necessary endpoints to MCP tools:
mcp := server.New(e)
// Include only specific endpoints
mcp.RegisterEndpoints([]string{
"/api/v1/users/:id",
"/api/v1/orders",
})
// Or exclude internal endpoints
mcp.ExcludeEndpoints([]string{
"/health", // Exclude health checks
})
For better control, register schemas manually:
type CreateUserRequest struct {
Name string `json:"name" jsonschema:"required,description=User full name"`
Email string `json:"email" jsonschema:"required,description=User email address"`
Age int `json:"age,omitempty" jsonschema:"minimum=0,maximum=150"`
}
type UserQuery struct {
Page int `form:"page,default=1" jsonschema:"minimum=1"`
Limit int `form:"limit,default=10" jsonschema:"maximum=100"`
Active bool `form:"active" jsonschema:"description=Filter by active status"`
}
mcp := server.New(e, &server.Config{BaseURL: "http://localhost:8080"})
// Register schemas for specific routes
mcp.RegisterSchema("POST", "/users", nil, CreateUserRequest{})
mcp.RegisterSchema("GET", "/users", UserQuery{}, nil)
Echo-MCP supports four schema generation approaches, with automatic fallback:
| Method | Use Case | Priority |
|---|---|---|
| Raw OpenAPI Schema | Using OpenAPI libraries like swaggest/openapi-go | First (if OpenAPISchema is set) |
| Swagger | Production APIs with Swaggo annotations | First (if EnableSwaggerSchemas is set) |
| Manual | Fine-grained control, complex validation | Second |
| Automatic | Quick prototyping, simple endpoints | Fallback |
// Option 1: Using raw OpenAPI schema (swaggest/openapi-go, etc.)
schema, _ := reflector.Spec.MarshalYAML()
mcp := server.New(e, &server.Config{
OpenAPISchema: string(schema), // Use raw schema
})
// Option 2: Using Swaggo
mcp := server.New(e, &server.Config{
EnableSwaggerSchemas: true, // Load from swaggo docs
})
// Option 3: Manual schemas for fine-grained control
mcp.RegisterSchema("POST", "/users", nil, CreateUserRequest{})
// Option 4: Automatic inference (fallback)
// No configuration needed - routes will use basic path/body inference
Once your server is running:
{
"mcpServers": {
"echo-api": {
"type": "http",
"url": "http://localhost:8080/mcp",
"timeout": 120
},
}
}
For local testing, use MCP Inspector:
npx @modelcontextprotocol/inspector http://localhost:8080/mcp
Install via CLI
npx mdskills install BrunoKrugel/echo-mcpMCP Wrapper for Echo Framework is a free, open-source AI agent skill. Wrap any existing Echo API into MCP tools, enabling AI agents to interact with your API through Model Context Protocol. Inspired by gin-mcp but for the Echo framework. - Zero Configuration: Works with any existing Echo API - Multiple Schema Sources: Support for Swaggo, raw OpenAPI YAML/JSON, and manual schemas - Filtering: Include/exclude endpoints with wildcard patterns - MCP Compatible: Works wi
Install MCP Wrapper for Echo Framework with a single command:
npx mdskills install BrunoKrugel/echo-mcpThis downloads the skill files into your project and your AI agent picks them up automatically.
MCP Wrapper for Echo Framework works with Claude Code, Claude Desktop, Cursor, Vscode Copilot, Windsurf, Continue Dev, Gemini Cli, Amp, Roo Code, Goose. Skills use the open SKILL.md format which is compatible with any AI coding agent that reads markdown instructions.