Skip to content

Latest commit

 

History

History
313 lines (224 loc) · 8.09 KB

krci-ai-cheatsheet.md

File metadata and controls

313 lines (224 loc) · 8.09 KB

KubeRocketAI CLI Cheatsheet

🚀 KubeRocketAI - AI-as-Code framework providing a centralized library of AI agents for software development workflows

🎯 Quick Start (3 Minutes)

Installation

# macOS (Recommended)
brew tap KubeRocketCI/homebrew-tap
brew install krci-ai

# Linux - Direct download
curl -L "https://github.com/KubeRocketCI/kuberocketai/releases/latest/download/krci-ai_Linux_x86_64.tar.gz" | tar -xz
chmod +x krci-ai && sudo mv krci-ai /usr/local/bin/

# Windows - Download from releases
# https://github.com/KubeRocketCI/kuberocketai/releases/latest

First Setup

# Install framework with IDE integration
krci-ai install --ide=claude    # For Claude Code
krci-ai install --ide=cursor    # For Cursor IDE
krci-ai install --all           # All IDE integrations

# Verify installation
krci-ai validate

📋 Command Reference

krci-ai install - Framework Installation

Install AI-as-Code framework components to your project.

Command Description
krci-ai install Install core framework components
krci-ai install --ide=cursor Install with Cursor IDE integration
krci-ai install --ide=claude Install with Claude Code integration
krci-ai install --ide=vscode Install with VS Code integration
krci-ai install --ide=windsurf Install with Windsurf IDE integration
krci-ai install --ide=all Install with all IDE integrations
krci-ai install --all Install core + all IDE integrations (shortcut)
krci-ai install --force Force installation (overwrite existing)

What Gets Installed:

  • .krci-ai/agents/ - 6 role-based agent definitions (PM, Architect, Developer, QA, BA, PO)
  • .krci-ai/tasks/ - Common workflow templates
  • .krci-ai/templates/ - Output formatting templates
  • .krci-ai/data/ - Reference data and standards
  • IDE-specific integration files (.cursor/rules/, .claude/commands/, etc.)

krci-ai list - Component Discovery

Discover available framework components.

Command Description
krci-ai list agents List all available agents with roles
krci-ai list agents -v List agents with dependency details

Available Agents:

  • pm - Senior Product Manager (Product strategy, requirements)
  • po - Senior Product Owner (User story creation, backlog)
  • ba - Senior Business Analyst (Requirements gathering)
  • architect - Senior Software Architect (System design, architecture)
  • dev - Software Developer (Implementation, code assistance)
  • qa - Senior QA Engineer (Testing strategy, quality assurance)

krci-ai validate - Framework Validation

Validate framework components and detect issues.

Command Description
krci-ai validate Standard validation with summary
krci-ai validate -v Verbose validation with detailed insights
krci-ai validate -q Quiet mode - minimal output

Validation Checks:

  • ✅ Agent YAML schema compliance
  • ✅ Task path link validation
  • ✅ Template file structure
  • ✅ Markdown links to framework files
  • ✅ Cross-platform file accessibility
  • ✅ Dependency analysis and circular dependency detection
  • ✅ Orphaned file detection

krci-ai bundle - Web Chat Integration

Generate agent bundles for web chat tools (ChatGPT, Gemini Pro, Claude Web).

Command Description
krci-ai bundle --all Generate complete bundle with all agents
krci-ai bundle --agent pm,architect Generate targeted bundle with specific agents
krci-ai bundle --agent pm --task create-prd Generate minimal bundle for specific task
krci-ai bundle --all --output my-bundle.md Custom output filename
krci-ai bundle --all --dry-run Preview bundle scope without generating

Bundle Contents:

  • All selected agent definitions with complete YAML
  • Agent dependencies, tasks, and templates
  • Project-specific context from .krci-ai/data/ and .krci-ai/templates/
  • System prompt structure with role definitions

krci-ai version - Version Information

Display version and build information.

krci-ai version
# Output: version, commit, build date, Go version, platform

krci-ai check-updates - Update Management

Check for available CLI updates.

krci-ai check-updates
# Queries GitHub API for latest releases

🎭 Agent Usage in IDEs

Claude Code Integration

After installing with --ide=claude:

/pm           # Activate Product Manager
/architect    # Activate Software Architect
/dev          # Activate Developer
/qa           # Activate QA Engineer
/ba           # Activate Business Analyst
/po           # Activate Product Owner

Agent Commands

Each agent provides these standard commands:

  • help - Show available commands
  • chat - (Default) Consultation and assistance
  • exit - Exit agent persona

Agent-Specific Commands:

  • PM: create-project-brief, create-prd
  • Architect: design-system, review-architecture
  • Developer: review, plan-implementation, implement
  • QA: create-test-plan, review-test-coverage

📁 Directory Structure

project-root/
├── .krci-ai/
│   ├── agents/          # 6 role-based agent definitions
│   │   ├── pm.yaml
│   │   ├── architect.yaml
│   │   ├── dev.yaml
│   │   ├── qa.yaml
│   │   ├── ba.yaml
│   │   └── po.yaml
│   ├── tasks/           # Common workflow templates
│   ├── templates/       # Output formatting templates
│   ├── data/           # Reference data and standards
│   └── bundle/         # Generated bundles (from bundle command)
├── .cursor/rules/      # Cursor IDE integration (if --ide=cursor)
├── .claude/commands/   # Claude Code integration (if --ide=claude)
└── .github/chatmodes/ # VS Code integration (if --ide=vscode)

🔧 Common Workflows

1. Project Setup

# Install and setup
krci-ai install --ide=claude
krci-ai validate

# List available agents
krci-ai list agents

2. Create Project Brief

# In Claude Code
/pm
create-project-brief
# Provide context about your project

3. Architecture Design

# In Claude Code
/architect
design-system
# Describe technical requirements

4. Implementation

# In Claude Code
/dev
implement
# Describe features to implement

5. Web Chat Bundle

# Generate bundle for ChatGPT/Claude Web
krci-ai bundle --all --output project-context.md
# Upload project-context.md to web chat tools

🛠️ Troubleshooting

Common Issues

# Permission denied
sudo krci-ai install --ide=cursor

# Validate installation
krci-ai validate -v

# Force reinstall
krci-ai install --force

# Check for updates
krci-ai check-updates

Validation Errors

  • Broken links: Check .krci-ai/ file paths in agent YAML
  • Missing tasks: Ensure task files exist in .krci-ai/tasks/
  • Schema errors: Validate agent YAML structure

IDE Integration Issues

  • Ensure IDE integration was installed: krci-ai install --ide=your-ide
  • Check for IDE-specific files in project root
  • Restart IDE after installation

📚 Additional Resources

  • Documentation: GitHub Repository
  • Issues & Support: GitHub Issues
  • Architecture: See docs/architecture.md in repository
  • Quick Start: See docs/quick-start.md in repository

🚀 Advanced Usage

Development Commands

# If you're developing KubeRocketAI itself
make build          # Build CLI binary
make test           # Run unit tests with coverage
make lint           # Run linting (golangci-lint)
make fmt            # Format code
make ci             # Run full CI pipeline locally

Schema Validation

KubeRocketAI uses JSON schemas for validation:

  • Agent YAML files follow strict schema definitions
  • Template and task files are validated for structure
  • Cross-references are verified for integrity