Agent

Agent is a lightweight, deployable system collector designed to run on any server or workstation. Its job is to gather hardware information, resource usage, and system-level metrics, and expose them to upstream services such as your Command Center.

The binary is intentionally small, dependency-minimal, and easy to deploy over SSH or via automated provisioning tools.

Features

• Collects hardware information (CPU, memory, disks, network)
• Monitors real-time resource usage
• Exposes information via gRPC endpoints
• Lightweight footprint, suitable for cloud VPS, on-prem servers, and embedded Linux systems
• Simple build and run workflow

Cloning

Inside the project we use a git submodule that refers to www.github.com/paradigm-ehb/agent-resources for the C-Library.

git clone --recursive <agent-repo-url>
# Or if already cloned:
git submodule update --init --recursive

Requirements

Before building the project or generating protobuf files, ensure the following tools are available in your environment:

• Go (compatible with your project version)
• protoc — Protocol Buffers compiler
• protoc-gen-go — Go protobuf generator
• protoc-gen-go-grpc — gRPC code generator for Go

Verify availability using:

protoc-gen-go --version
protoc-gen-go-grpc --version

If they’re not detected in your PATH, install them using your preferred method (Go install, Homebrew, package managers, etc.).

Building Protobuf Files

The project includes a helper script:

./build.sh

This script regenerates all .pb.go and gRPC service definitions. Use it after editing any .proto file under internal/connection.

Running the Agent

The agent can be run directly during development:

go run cmd/main.go

The default configuration starts the gRPC server on port 50051. You can override this using command-line flags or environment variables depending on your setup.

Project Structure

A minimal overview of the source layout:

cmd/
  main.go          Entry point for the agent  
internal/
  connection/      Protobuf definition and generated gRPC files
  system/          System info collectors (platform-specific)
  handlers/        gRPC service implementations
build.sh           Regenerate protobuf files
README.md          You are here

This structure keeps public API code clean while isolating platform-specific logic inside internal/.

Usage Example (Command Line)

Tools such as grpcurl allow quick manual interaction with the agent:

grpcurl -plaintext localhost:50051 agent.Greeter/SayHello -d '{"name":"Appie"}'

Useful during debugging or when testing connectivity between Agent and Command Center.