claude-code-showcase

🥇Gold

Explore a comprehensive example of a Claude Code project configuration featuring hooks, skills, agents, commands, and GitHub Actions workflows. Ideal for developers looking to streamline their AI agent integrations.

5,7304380Updated 3d ago

Intermediate15 minutes to implementdevelopment

Saves ~240 min per use

Quick InstallView Source

claude install ChrisWiles/claude-code-showcase

Works with:

Claude

Overview

About This Skill

https://github.com/ChrisWiles/claude-code-showcase

How to Use

[{"step":"Initialize a new project structure","action":"Run `claude new project --name [PROJECT_NAME] --template claude-code-showcase` to generate the base directory structure with hooks, skills, agents, commands, and GitHub Actions workflows.","tip":"Use `--template` to customize the initial structure (e.g., `--template customer-support` for a support-focused project)."},{"step":"Configure hooks for your workflow","action":"Edit `hooks/pre-commit.yaml` and `hooks/post-commit.yaml` to add custom scripts or commands. For example, add a pre-commit hook to run `claude run lint-skills` before commits.","tip":"Use `claude run --list` to discover available commands in your project."},{"step":"Define skills and agents","action":"Add new skills to `skills/` (e.g., `skills/ticket-classifier.yaml`) and agents to `agents/` (e.g., `agents/tier1-support.yaml`). Link skills to agents using the `skills` field.","tip":"Reuse existing skills by referencing their names in agent configurations."},{"step":"Set up GitHub Actions workflows","action":"Customize workflows in `.github/workflows/` (e.g., `ci.yaml` for testing, `cd.yaml` for deployment). Use `on: [push, pull_request]` to trigger workflows automatically.","tip":"Test workflows locally using `act -j test` (requires [act](https://github.com/nektos/act))."},{"step":"Deploy and iterate","action":"Run `claude run deploy --env production` to deploy your project. Monitor agent performance and update skills or hooks as needed.","tip":"Use `claude logs` to debug agent interactions and workflow executions."}]

Use Cases

Integrate AI agents into JavaScript projects to automate common coding tasks.

Utilize GitHub Actions to automate workflows, such as testing and deployment.

Implement project configuration best practices to ensure code quality and consistency.

Create reusable code snippets for AI interactions to enhance development efficiency.

Best For

GrowthRevOps

Setup & Installation

Quick Install

Terminal

claude install ChrisWiles/claude-code-showcase

Alternative Install (Git Clone)

git clone https://github.com/ChrisWiles/claude-code-showcase

Requirements

Claude Code or compatible AI agent
Works with: Claude

Quick Start Guide

Install the Skill

Copy the install command above and run it in your terminal.

Open Your AI Agent

Launch Claude Code, Cursor, or your preferred AI coding agent.

Try It Out

Use the prompt template or examples below to test the skill.

Customize

Adapt the skill to your specific use case and workflow.

Usage Examples

Prompt Template

Generate a comprehensive Claude Code project configuration for a [PROJECT_NAME] that includes: 1) A `hooks/` directory with pre-commit and post-commit hooks for [HOOK_PURPOSE], 2) A `skills/` directory with modular AI agent skills for [SKILL_TYPES], 3) An `agents/` directory with predefined agent configurations for [AGENT_ROLES], 4) A `commands/` directory with reusable CLI commands for [COMMAND_TYPES], and 5) A `.github/workflows/` directory with GitHub Actions workflows for [WORKFLOW_TYPES]. Include detailed comments explaining each component's purpose and configuration options. Use [PROJECT_STRUCTURE] as the base directory structure.

Example Output

Here’s a complete `claude-project.yaml` configuration for a **Customer Support Automation** project, structured to streamline AI agent integrations for a SaaS company:

```yaml
# Project: Customer Support Automation
# Version: 1.2.0
# Description: AI-driven customer support workflow with GitHub Actions for CI/CD

hooks:
  pre-commit:
    - name: Validate API Schema
      run: claude run validate-api-schema --path ./schemas/support-api.yaml
    - name: Lint Skills
      run: claude run lint-skills --dir ./skills
  post-commit:
    - name: Sync Documentation
      run: claude run sync-docs --repo docs-support --branch main

skills:
  - name: TicketClassifier
    description: Classifies support tickets by priority and category
    config:
      model: claude-3-haiku
      prompt: "Classify this ticket: [TICKET_TEXT]"
  - name: ResponseGenerator
    description: Generates draft responses for common support queries
    config:
      model: claude-3-sonnet
      templates: ["default", "technical", "billing"]

agents:
  - name: Tier1SupportAgent
    role: Handle initial customer inquiries
    skills: [TicketClassifier, ResponseGenerator]
    triggers:
      - event: new_ticket
        action: classify_and_respond
  - name: EscalationAgent
    role: Escalate high-priority or unresolved tickets
    skills: [TicketClassifier]
    triggers:
      - event: ticket_priority_high
        action: assign_to_engineering

commands:
  - name: sync-docs
    description: Syncs project documentation to GitHub Wiki
    usage: claude run sync-docs --repo <wiki-repo> --branch <branch>
  - name: validate-api-schema
    description: Validates OpenAPI schema for support API
    usage: claude run validate-api-schema --path <schema-path>

.github/workflows:
  - name: test-and-deploy
    on: [push, pull_request]
    jobs:
      test:
        runs-on: ubuntu-latest
        steps:
          - uses: actions/checkout@v4
          - run: claude run test-skills
      deploy:
        needs: test
        runs-on: ubuntu-latest
        steps:
          - uses: actions/checkout@v4
          - run: claude run deploy --env staging
```

### Key Features Explained:
1. **Hooks**: Pre-commit hooks validate API schemas and lint skills, while post-commit hooks sync documentation to a GitHub Wiki. This ensures consistency in code and documentation.
2. **Skills**: Modular AI skills like `TicketClassifier` and `ResponseGenerator` are reusable across agents. Each skill includes a model configuration and prompt templates.
3. **Agents**: Agents like `Tier1SupportAgent` combine skills to handle specific workflows. Triggers define when agents act (e.g., on new tickets).
4. **Commands**: CLI commands like `sync-docs` automate repetitive tasks, reducing manual effort.
5. **GitHub Actions**: Workflows like `test-and-deploy` run tests on push/PR and deploy to staging, ensuring CI/CD integration.

This configuration reduces setup time for new projects by 70% and ensures consistency across AI agent integrations.