docs: update README with new import and discussion

This commit is contained in:
Toby Padilla 2025-05-05 14:27:17 -06:00 • committed by Toby Padilla
commit 073555ba04
2 changed files with 29 additions and 83 deletions

2
.gitignore vendored
View file

@ -3,3 +3,5 @@ data
.env .env
.mcpregistry* .mcpregistry*
**/bin **/bin
cmd/registry/registry
registry

106
README.md
View file

@ -1,8 +1,10 @@
# MCP Registry # MCP Registry
A reference implementation of a registry service for Model Context Protocol (MCP) servers. A community driven registry service for Model Context Protocol (MCP) servers.
> **Note:** This project is currently a work in progress. ## Development Status
This project is being built in the open and is currently in the early stages of development. Please see the [overview discussion](https://github.com/modelcontextprotocol/registry/discussions/11) for the project scope and goals. If you would like to contribute, please check out the [contributing guidelines](CONTRIBUTING.md).
## Overview ## Overview
@ -18,55 +20,17 @@ The MCP Registry service provides a centralized repository for MCP server entrie
- Comprehensive API documentation - Comprehensive API documentation
- Pagination support for listing registry entries - Pagination support for listing registry entries
## Project Structure
```
├── cmd/ # Application entry points
├── config/ # Configuration files
├── internal/ # Private application code
│ ├── api/ # HTTP server and request handlers
│ ├── config/ # Configuration management
│ ├── model/ # Data models
│ └── service/ # Business logic
├── pkg/ # Public libraries
├── scripts/ # Utility scripts
├── tools/ # Command line tools
│ └── importer/ # MongoDB data importer tool
└── build/ # Build artifacts
```
## Getting Started ## Getting Started
### Prerequisites ### Prerequisites
- Go 1.18 or later - Go 1.18 or later
- MongoDB (optional, for production use) - MongoDB
- Docker (optional, for containerized deployment) - Docker (optional, but recommended for development)
### Building ## Running
```bash The easiest way to get the registry running is to use `docker compose`. This will setup the MCP Registry service, import the seed data and run MongoDB in a local Docker environment.
make build
```
This will create the `registry` binary in the `build/` directory.
### Running
```bash
./build/mcp-registry
```
Alternatively, run
```bash
MCP_REGISTRY_DATABASE_URL="mongodb://localhost:27017" ./build/registry
```
By default, the service will run on `http://localhost:8080`.
### Docker Deployment
You can build and run the service using Docker:
```bash ```bash
# Build the Docker image # Build the Docker image
@ -78,52 +42,32 @@ docker compose up
This will start the MCP Registry service and MongoDB with Docker, exposing it on port 8080. This will start the MCP Registry service and MongoDB with Docker, exposing it on port 8080.
### Using the Fake Service ## Building
For development and testing purposes, the application includes a fake service with pre-populated registry entries. To use the fake service: If you prefer to run the service locally without Docker, you can build and run it directly using Go.
1. Set the environment to "test":
```bash ```bash
export MCP_REGISTRY_ENVIRONMENT=test # Build a registry executable
./build/registry go build ./cmd/registry
``` ```
This will create the `registry` binary in the current directory. You'll need to have MongoDB running locally or with Docker.
Alternatively, run By default, the service will run on `http://localhost:8080`.
## Project Structure
```bash
make run-test
``` ```
├── cmd/ # Application entry points
The fake service provides three sample MCP registry entries with the following capabilities: ├── config/ # Configuration files
├── internal/ # Private application code
- Registry 1: Code generation and completion capabilities │ ├── api/ # HTTP server and request handlers
- Registry 2: Chat and knowledge base capabilities │ ├── config/ # Configuration management
- Registry 3: Data visualization and analysis capabilities │ ├── model/ # Data models
│ └── service/ # Business logic
You can interact with the fake data through the API endpoints: ├── pkg/ # Public libraries
└── scripts/ # Utility scripts
- List all entries: `GET /servers`
The fake service is useful for:
- Frontend development without a real backend
- Testing API integrations
- Example data structure reference
## Tools
### Data Importer
A command-line tool for importing server data from a JSON file into a MongoDB database:
```bash
cd tools/importer
go run main.go -uri mongodb://localhost:27017 -db mcp_registry -collection servers -seed ../../data/seed.json
``` ```
For more details on the importer tool, see the [importer README](./tools/importer/README.md).
## API Documentation ## API Documentation
The API is documented using Swagger/OpenAPI. You can access the interactive Swagger UI at: The API is documented using Swagger/OpenAPI. You can access the interactive Swagger UI at: