Skip to content

Latest commit

 

History

History
309 lines (222 loc) · 5.84 KB

INSTALLATION.md

File metadata and controls

309 lines (222 loc) · 5.84 KB

Installation Guide

Prerequisites

Required

  • Clawdbot v1.0.0 or later
  • Node.js v18+ (usually installed with Clawdbot)
  • kubectl (for Kubernetes operations)

Optional (depending on which skills you use)

  • aws CLI - For AWS operations
  • terraform - For infrastructure as code
  • docker - For container operations
  • argocd CLI - For GitOps operations
  • gh CLI - For GitHub operations

Installation

Method 1: Install from GitHub (Recommended)

# Clone the repository
git clone https://github.com/clawdbot/devops-execution-engine
cd devops-execution-engine

# Install as Clawdbot skill
clawdbot skills:install .

# Verify installation
clawdbot skills:list | grep devops-execution-engine

Method 2: Install from ClawdHub (When Available)

# Search for the skill
clawdbot skills:search devops

# Install directly
clawdbot skills:install devops-execution-engine

# Verify
clawdbot skills:list

Method 3: Manual Installation

# Clone the repo
git clone https://github.com/clawdbot/devops-execution-engine

# Copy to Clawdbot skills directory
cp -r devops-execution-engine ~/.clawdbot/skills/

# Restart Clawdbot
clawdbot gateway restart

Configuration

Basic Configuration

Create ~/.clawdbot/skills/devops-execution-engine/config.yaml:

# Execution engine config
execution:
  auto_approve_low_risk: false    # Auto-approve LOW risk actions
  pause_between_steps: true       # Pause after each step
  timeout_default: 300            # Default timeout (seconds)
  dry_run: false                  # Test mode (no actual execution)
  
# Logging
audit:
  log_path: "memory/actions-log.jsonl"
  log_level: "info"              # debug|info|warn|error
  
# Safety
safety:
  require_approval: true          # Always require approval
  allow_critical: false           # Block CRITICAL risk actions
  block_production: false         # Extra safety for prod namespaces
  
# Skills to enable (leave empty to enable all)
enabled_skills: []

Kubernetes Configuration

Ensure your kubectl is configured with the correct context:

# Check current context
kubectl config current-context

# List available contexts
kubectl config get-contexts

# Switch context if needed
kubectl config use-context <context-name>

AWS Configuration (Optional)

If using AWS skills:

# Configure AWS CLI
aws configure

# Or use environment variables
export AWS_PROFILE=your-profile
export AWS_REGION=us-east-1

Read-Only Access (Recommended)

For safety, start with read-only Kubernetes access:

# read-only-sa.yaml
apiVersion: v1
kind: ServiceAccount
metadata:
  name: clawd-readonly
  namespace: default
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
  name: clawd-readonly-binding
subjects:
- kind: ServiceAccount
  name: clawd-readonly
  namespace: default
roleRef:
  kind: ClusterRole
  name: view  # Built-in read-only role
  apiGroup: rbac.authorization.k8s.io

Apply and get token:

kubectl apply -f read-only-sa.yaml

# Get token (k8s 1.24+)
kubectl create token clawd-readonly --duration=87600h

Update your kubeconfig to use this token.

Verification

Test Basic Functionality

# Start Clawdbot chat
clawdbot chat

# Test commands:
"List all skills"
"Use devops-execution-engine"
"Check system health"

Test Kubernetes Integration

"Check cluster health"
"List all pods"
"Show recent events"

You should see diagnostic output without errors.

Test Plan Generation

"Generate a plan to scale api-service to 5 replicas"

You should see a structured execution plan created.

Directory Structure After Installation

~/.clawdbot/skills/devops-execution-engine/
├── SKILL.md
├── LICENSE
├── README.md
├── config.yaml                 # Your configuration
├── core/                       # Execution engine
├── templates/                  # Plan templates
├── skills/                     # 11 DevOps skills
├── examples/                   # Example plans
└── docs/                       # Documentation

~/clawd/memory/                 # Created in your workspace
├── execution-plans/            # Generated plans
└── actions-log.jsonl           # Audit trail

Troubleshooting

Skill Not Found

# Check installation
clawdbot skills:list

# Reinstall if needed
cd devops-execution-engine
clawdbot skills:install .

Permission Errors

# Check directory permissions
ls -la ~/.clawdbot/skills/

# Fix if needed
chmod -R 755 ~/.clawdbot/skills/devops-execution-engine

kubectl Not Working

# Verify kubectl is in PATH
which kubectl

# Check context
kubectl config current-context

# Test basic command
kubectl get nodes

Plans Not Being Created

# Check memory directory exists
ls ~/clawd/memory/execution-plans/

# Create if needed
mkdir -p ~/clawd/memory/execution-plans

Upgrading

# Pull latest changes
cd devops-execution-engine
git pull origin main

# Reinstall
clawdbot skills:install .

# Restart Clawdbot
clawdbot gateway restart

Uninstalling

# Remove skill
rm -rf ~/.clawdbot/skills/devops-execution-engine

# Restart Clawdbot
clawdbot gateway restart

# Optional: Keep or remove execution history
# rm -rf ~/clawd/memory/execution-plans
# rm ~/clawd/memory/actions-log.jsonl

Next Steps

  • Read SKILLS.md for available skills
  • Read SAFETY.md for safety model
  • See EXAMPLES.md for usage examples
  • Start with simple read-only operations
  • Build trust before enabling execution

Getting Help