project-starter

A universal Claude Code workflow plugin with specialized agents, skills, hooks, and output styles for any software project.

---

Quick Start

Option 1: CLI (Per-Session)

# Clone the plugin
git clone https://github.com/CloudAI-X/claude-workflow.git

# Run Claude Code with the plugin
claude --plugin-dir ./claude-workflow

Option 2: Agent SDK

import { query } from "@anthropic-ai/claude-agent-sdk";

for await (const message of query({
  prompt: "Hello",
  options: {
    plugins: [{ type: "local", path: "./claude-workflow" }],
  },
})) {
  // Plugin commands, agents, and skills are now available
}

Option 3: Install Permanently

# Install from marketplace (when available)
claude plugin install project-starter

# Or install from local directory
claude plugin install ./claude-workflow

Verify Installation

After loading the plugin, verify it's working:

> /plugin

Tab to Installed - you should see project-starter listed. Tab to Errors - should be empty (no errors).

These commands become available:

/project-starter:architect    # Architecture-first mode
/project-starter:rapid        # Ship fast mode
/project-starter:commit       # Auto-generate commit message
/project-starter:verify-changes  # Multi-agent verification

---

What's Included

| Component | Count | Description | | ------------ | ----- | ------------------------------------------------------------------------ | | Agents | 7 | Specialized subagents for code review, debugging, security, etc. | | Commands | 17 | Slash commands for workflows and output styles | | Skills | 10 | Knowledge domains Claude uses autonomously | | Hooks | 9 | Automation scripts for formatting, security, verification, notifications |

---

Usage Examples

Commands in Action

Auto-commit your changes:

> /project-starter:commit

Looking at staged changes...
✓ Created commit: feat(auth): add JWT refresh token endpoint

Full git workflow:

> /project-starter:commit-push-pr

✓ Committed: feat: add user dashboard
✓ Pushed to origin/feature/dashboard
✓ Created PR #42: https://github.com/you/repo/pull/42

Verify before shipping:

> /project-starter:verify-changes

Spawning verification agents...
├─ build-validator: ✓ Build passes
├─ test-runner: ✓ 42 tests pass
├─ lint-checker: ⚠ 2 warnings (non-blocking)
└─ security-scanner: ✓ No vulnerabilities

Ready to ship!

Agents in Action

Agents spawn automatically based on your request:

You say: "The login is broken, users get 401 errors"

[debugger agent activated]
→ Checking auth middleware... found issue
→ Token validation uses wrong secret in production
→ Fix: Update AUTH_SECRET in .env.production

You say: "Review my changes"

[code-reviewer agent activated]
→ Analyzing 3 files changed...
✓ Logic is correct
⚠ Missing null check on line 42
⚠ Consider adding rate limiting to this endpoint

You say: "Add authentication to the API"

[orchestrator agent activated]
→ Breaking down into subtasks:
  1. Design auth schema (spawning architect)
  2. Implement JWT middleware
  3. Add login/register endpoints
  4. Write tests (spawning test-architect)
  5. Update API docs (spawning docs-writer)

Skills in Action

Skills provide domain knowledge automatically:

You ask: "How should I structure the payment service?"

[designing-architecture skill applied]
→ Recommending hexagonal architecture
→ Payment providers as adapters
→ Core domain isolated from infrastructure

You ask: "Make this endpoint faster"

[optimizing-performance skill applied]
→ Adding database indexes
→ Implementing response caching
→ Using pagination for large results

Hooks in Action

Hooks run automatically on events:

Security block (pre-edit):

⛔ BLOCKED: Potential secret detected
   File: src/config.ts, Line 5
   Pattern: API key (sk-...)

   Remove the secret and use environment variables.

Auto-format (post-edit):

✓ Formatted with prettier: src/components/Button.tsx
✓ Formatted with black: scripts/deploy.py

Desktop notifications:

🔔 "Claude needs input" - when waiting for your response
🔔 "Task complete" - when finished

---

Commands Reference

All commands use the format /project-starter:<command>.

Output Styles

| Command | Mode | | ---------------------------- | --------------------------------------------- | | /project-starter:architect | System design mode - architecture before code | | /project-starter:rapid | Fast development - ship quickly, iterate | | /project-starter:mentor | Teaching mode - explain the "why" | | /project-starter:review | Code review mode - strict quality |

Git Workflow (Inner-Loop)

| Command | Purpose | | ------------------------------------ | ----------------------------------------- | | /project-starter:commit | Auto-generate conventional commit message | | /project-starter:commit-push-pr | Commit → Push → Create PR (full workflow) | | /project-starter:quick-fix | Fast fix for lint/type errors | | /project-starter:add-tests | Generate tests for recent changes | | /project-starter:lint-fix | Auto-fix all linting issues | | /project-starter:sync-branch | Sync with main (rebase or merge) | | /project-starter:summarize-changes | Generate standup/PR summaries |

Verification

| Command | Purpose | | ---------------------------------- | --------------------------------------- | | /project-starter:verify-changes | Multi-subagent adversarial verification | | /project-starter:validate-build | Build process validation | | /project-starter:run-tests | Tiered test execution | | /project-starter:lint-check | Code quality checks | | /project-starter:security-scan | Security vulnerability detection | | /project-starter:code-simplifier | Post-implementation cleanup |

---

Agents

Agents are specialized subagents that Claude spawns automatically based on your task.

| Agent | Purpose | Auto-Triggers | | ------------------ | -------------------------------- | ------------------------------------------- | | orchestrator | Coordinate multi-step tasks | "improve", "refactor", multi-module changes | | code-reviewer | Review code quality | After code changes, before commits | | debugger | Systematic bug investigation | Errors, test failures, crashes | | docs-writer | Technical documentation | README, API docs, guides | | security-auditor | Security vulnerability detection | Auth, user input, sensitive data | | refactorer | Code structure improvements | Technical debt, cleanup | | test-architect | Design test strategies | Adding/improving tests |

---

Skills

Skills are knowledge domains that Claude uses autonomously when relevant.

| Skill | Domain | | ----------------------------- | ----------------------------------------------------- | | analyzing-projects | Understand codebase structure and patterns | | designing-tests | Unit, integration, E2E test approaches | | designing-architecture | Clean Architecture, Hexagonal, etc. | | optimizing-performance | Speed up applications, identify bottlenecks | | managing-git | Version control, conventional commits | | designing-apis | REST/GraphQL patterns and best practices | | parallel-execution | Multi-subagent parallel task execution patterns | | web-design-guidelines | UI review against Vercel Web Interface Guidelines | | vercel-react-best-practices | React/Next.js performance optimization (45 rules) | | convex-backend | Convex backend development (functions, schemas, etc.) |

---

Hooks

Hooks run automatically on specific events.

| Hook | Trigger | Action | | --------------------- | ------------- | -------------------------------------- | | Security scan | Edit/Write | Blocks commits with potential secrets | | File protection | Edit/Write | Blocks edits to lock files, .env, .git | | Auto-format | Edit/Write | Runs prettier/black/gofmt by file type | | Command logging | Bash | Logs to .claude/command-history.log | | Environment check | Session start | Validates Node.js, Python, Git | | Prompt analysis | User prompt | Suggests appropriate agents | | Auto-verify | Task complete | Runs tests/lint, reports results | | Input notification | Input needed | Desktop notification | | Complete notification | Task complete | Des