title: "Tools & Integrations" description: "CLI tools, editor plugins, CI/CD connectors, and third-party services that extend AI coding workflows" section: "Ecosystem" readTime: "10 min"

Tools & Integrations

The AI coding ecosystem extends well beyond the core tools. This page catalogs the integrations, plugins, and CLI companions that make Claude Code, GitHub Copilot, and Cursor more powerful in real development workflows.

CLI Companions

Claude Code + Git Hooks

Automate Claude Code in your git workflow:

# .husky/pre-commit — AI security scan before every commit
npx claude -p "Review this staged diff for hardcoded secrets and OWASP vulnerabilities.
Output PASS or BLOCK: <reason> only." <<< "$(git diff --cached)"

# .husky/commit-msg — AI commit message improvement
ORIGINAL=$(cat "$1")
IMPROVED=$(npx claude -p "Improve this commit message to follow conventional commit format. 
Return only the improved message: $ORIGINAL")
echo "$IMPROVED" > "$1"

GitHub CLI + Copilot Extension

The gh CLI with GitHub Copilot extension enables AI from the terminal:

# Install
gh extension install github/gh-copilot
 
# Usage
gh copilot suggest "undo last pushed commit"
gh copilot explain "git rebase -i HEAD~3"

aider (Open Source AI Pair Programmer)

Aider works with any OpenAI/Anthropic/local model and integrates with your git history:

pip install aider-chat
 
# Use with Claude
aider --model claude-3-5-sonnet-20241022 src/auth.py
 
# Use with GPT-4o
aider --model gpt-4o src/auth.py

Key difference from Claude Code: aider uses a map of your entire repo's symbols for context, which can be more accurate for large codebases.

VS Code Extensions

Extension	Purpose	Install
GitHub Copilot	Core AI completions + chat	Default
GitHub Copilot Chat	Chat interface, agent mode	Default with Copilot
Error Lens	Inline error display — improves AI diagnostic context	`usernamehw.errorlens`
REST Client	Test APIs inline — AI can read `.http` files	`humao.rest-client`
GitLens	Rich git history — Copilot reads blame context	`eamodio.gitlens`
Todo Tree	Track TODO/FIXME — useful for AI task planning	`gruntfuggly.todo-tree`

CI/CD Integrations

GitHub Actions: Copilot Code Review

# .github/workflows/copilot-review.yml
name: Copilot Review
on:
  pull_request:
    types: [opened, synchronize]
 
jobs:
  review:
    runs-on: ubuntu-latest
    permissions:
      contents: read
      pull-requests: write
    steps:
      - uses: actions/checkout@v4
      - name: Copilot Review
        uses: github/copilot-pr-review-action@v1
        with:
          github-token: ${{ secrets.GITHUB_TOKEN }}

GitHub Actions: Claude Code Security Scan

name: AI Security Scan
on: [pull_request]
 
jobs:
  security:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
        with:
          fetch-depth: 0
      - uses: actions/setup-node@v4
        with:
          node-version: '20'
      - run: npm install -g @anthropic-ai/claude-code
      - name: Scan changed files
        env:
          ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
        run: |
          git diff origin/main...HEAD --name-only | \
          xargs -I {} claude -p "Security scan for OWASP Top 10 vulnerabilities in {}"

Pre-commit Framework

# .pre-commit-config.yaml
repos:
  - repo: local
    hooks:
      - id: ai-review
        name: AI Code Review
        entry: bash -c 'git diff --cached | claude -p "Flag only critical bugs and security issues. Output PASS or list issues."'
        language: system
        pass_filenames: false

Testing Integrations

Vitest + AI Coverage Analysis

# Generate coverage report, ask AI to analyze gaps
npx vitest run --coverage
cat coverage/coverage-summary.json | claude -p "Identify the 5 most critical gaps 
in test coverage based on this report. Suggest specific test cases for each."

Playwright + Claude Vision

Claude Code with Playwright MCP server for end-to-end testing:

claude "Use Playwright to test the checkout flow on localhost:3000.
Navigate to a product page, add to cart, proceed to checkout, 
verify the order summary shows the correct total.
Take a screenshot at each major step and flag any failures."

Documentation Generation

TypeDoc + AI Enhancement

# Generate base docs
npx typedoc --out docs/api src/
 
# Enhance with AI
claude "Read docs/api/index.html and improve all JSDoc descriptions 
that are just auto-generated stubs. Focus on functions in src/lib/."

Storybook + Copilot Story Generation

In VS Code with Storybook open:

@workspace Generate Storybook stories for all components in src/components/ 
that don't have a .stories.tsx file. Include default, loading, and error states.

Productivity Tools

Raycast + Claude

Raycast (macOS) has a Claude extension for instant AI access outside VS Code:

# Quick refactor: select code → Raycast → "Refactor with Claude"
# Explain: select error → Raycast → "Explain this error"

Warp Terminal

Warp terminal has built-in AI command suggestions — combines well with Claude Code:

Ctrl+~ for AI command suggestions
Natural language → shell command
AI-generated command explanations in line

Monitoring AI Code Quality

Track the output quality of your AI tools over time:

Metric	How to Measure
AI churn rate	% of AI-generated lines modified within 1 week (via `git blame`)
Test flake rate	Tests generated by AI that are intermittently failing
Security finding rate	Findings per PR from security scan CI step
Completion acceptance rate	Copilot telemetry in VS Code settings