@llm-dev-ops/llm-schema-registry-server

LLM Schema Registry Server - HTTP/gRPC server for schema management

Installation

npm install -g @llm-dev-ops/llm-schema-registry-server

Usage

# Start the server with default configuration
llm-schema-server

# Start with custom configuration
llm-schema-server --config /path/to/config.yaml

# Start on a specific port
llm-schema-server --port 8080

# View help
llm-schema-server --help

Configuration

The server can be configured using a YAML configuration file. Example:

server:
  http_port: 8080
  grpc_port: 50051
  host: "0.0.0.0"

database:
  postgres:
    url: "postgresql://user:password@localhost/schema_registry"
    max_connections: 10

storage:
  type: "postgres"  # or "s3"

cache:
  enabled: true
  ttl_seconds: 300

security:
  auth_enabled: false
  jwt_secret: "your-secret-key"

Features

• HTTP REST API for schema management
• gRPC API for high-performance access
• PostgreSQL/S3 storage backends
• Built-in caching with Redis support
• JWT authentication
• Prometheus metrics
• OpenTelemetry tracing

API Endpoints

HTTP REST API (default port 8080)

• POST /schemas - Register a new schema
• GET /schemas/:namespace/:name/:version - Get a schema
• POST /schemas/validate - Validate data against a schema
• POST /schemas/compatibility - Check schema compatibility
• GET /schemas/:namespace - List schemas in a namespace
• DELETE /schemas/:namespace/:name/:version - Delete a schema

gRPC API (default port 50051)

See the API documentation for gRPC service definitions.

Documentation

For more information, visit: https://github.com/globalbusinessadvisors/llm-schema-registry

License

Apache-2.0