Skip to content

CLAUDE.md

Latest commit

 

History

History
49 lines (36 loc) · 2.17 KB

CLAUDE.md

File metadata and controls

49 lines (36 loc) · 2.17 KB

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Project Overview

The AMRC Connectivity Stack (ACS) is a Kubernetes-deployed implementation of the Factory+ framework for industrial connectivity and data management. It consists of central cluster services (MQTT broker, Directory, Auth, ConfigDB, Historians, Manager UI) and edge agents that collect data from industrial devices.

Repository Structure

  • lib/ - Shared libraries (must build first)

    • js-service-client - Main client library for Factory+ services
    • js-service-api - Base classes for building service APIs
    • js-edge-driver - Base driver class for edge translators
    • js-sparkplug-app - Sparkplug B protocol utilities
    • js-pg-client, js-rx-client, js-rx-util - Database and reactive utilities
    • py-edge-driver - Python edge driver base
    • java-service-client - Java client library

  • acs-* - Central cluster services (Auth, ConfigDB, Directory, etc.)

  • edge-* - Edge protocol translators (Modbus, BACnet, ADS, etc.)

  • historian-*, uns-ingester-* - Data ingestion services

  • deploy/ - Helm chart for Kubernetes deployment

  • mk/ - Makefile fragments

JavaScript Services

Most services are ES modules using @amrc-factoryplus/service-client for Factory+ integration. Services reference local libraries via file:../lib/js-* in package.json.

TypeScript services (like acs-edge) use:

npm run dev    # Development with ts-node-dev
npm run build  # Compile TypeScript
npm run test   # Run Jest tests

Key Patterns

  • Services authenticate via Kerberos to the MQTT broker
  • Configuration is stored in ConfigDB and accessed via the service client
  • Edge agents publish Sparkplug B messages to the MQTT broker
  • The @amrc-factoryplus/service-client library provides ServiceClient class for accessing Factory+ services

Contributing

  • Branch naming: initials/branch-description or feature/xxx for long-running branches
  • Commit messages: imperative mood, explain the "why", reference issues
  • Keep PRs focused on a single issue/feature
  • Rebase onto main rather than merging