doover-development

Doover Skills

A collection of Claude Skills for developing applications on the Doover platform.

Installation

Add this plugin to Claude Code:

/plugin marketplace add doover/doover-skills

Available Skills

doover-app-workflow

End-to-end workflow for creating new Doover apps:

• Research existing apps for integration
• Create GitHub repo from template
• Configure organization and registry
• Implement application logic
• Test with local simulators
• Publish to Doover platform

Start here when building a new app.

doover-device-apps

Comprehensive guide for creating and developing Doover device applications. Covers:

• Application lifecycle and structure
• Configuration schemas
• UI components
• Simulators for local testing
• Platform interface for hardware I/O
• Best practices

doover-cloud-apps

Guide for processors and integrations (cloud-based apps):

• Processor apps (channel and schedule triggered)
• Integration apps (HTTP ingestion endpoints)
• Event handlers (on_message_create, on_schedule, on_ingestion_endpoint)
• Cloud app configuration
• Integration + Processor patterns

doover-cli

Complete reference for the Doover CLI:

• doover app create - Create new applications
• doover app run - Local development
• doover app build - Build Docker images
• doover app publish - Deploy to platform
• doover app test/lint/format - Code quality
• doover app channels - Debug data streams

doover-data-management

Doover's channel-based data architecture:

• Channels as the foundation (publish/subscribe)
• Tags system (tag_values channel)
• UI system (ui_state, ui_cmds channels)
• Inter-agent communication
• Alerts and notifications

pydoover

API reference for the pydoover Python library:

• Application class and lifecycle
• Configuration schema types
• UI components (Variables, Parameters, Actions)
• Hardware interfaces (Platform, Modbus)
• State machine support
• Utility functions

doover-app-explorer

Search and explore existing Doover apps:

• Browse apps at https://admin.doover.com/app-explorer
• Query apps via API
• Find apps by functionality or dependencies
• Evaluate apps for integration
• Understand config schemas and app references

doover-admin

Platform administration:

• Organization setup
• Device types
• Deployment configuration
• Monitoring and health checks

Quick Start

1. Install the plugin in Claude Code
2. Describe what you want to build:
   • "I need a Doover app that pushes sensor data to Power BI"
   • "Create an app to control a pump based on tank level"
   • "Build an integration to receive webhooks from my fleet management system"
3. The doover-app-workflow skill will guide you through:
   • Searching for existing apps
   • Setting up a new project from the template
   • Implementing and testing
   • Publishing to Doover

Usage

Invoke skills in Claude Code by asking questions about Doover development:

• "How do I create a new Doover app?"
• "Show me how to use state machines"
• "What CLI commands are available?"
• "How do I publish data to channels?"
• "What UI components are available in pydoover?"
• "Find apps that work with Modbus"
• "I need a Doover app that pushes tag data to Power BI"

CI/CD

This repository uses GitHub Actions for continuous integration and deployment.

Validation Workflow

Runs on every push and pull request to main:

• Validates marketplace plugin structure using claude plugin validate
• Checks JSON syntax for all JSON files
• Verifies all referenced skill files exist

Release Workflow

Runs when a version tag is pushed (e.g., v1.0.0):

• Validates the marketplace structure
• Verifies the tag version matches plugin.json
• Creates a GitHub Release with installation instructions

Creating a Release

1. Update the version in .claude-plugin/plugin.json
2. Commit and push to main
3. Create and push a version tag:

   git tag v1.0.0
   git push origin v1.0.0
   

Resources

• Doover Platform
• pydoover Documentation

License

MIT License - see LICENSE for details.

Support

For support, contact [email protected]