|
| 1 | +# Awesome Claude Agents |
| 2 | + |
| 3 | +A collection of **24 specialized AI sub-agents** for [Claude Code](https://claude.ai/code) that collaborate as a full software development team. Instead of relying on a single generalist model, this system routes every task to the right expert automatically. |
| 4 | + |
| 5 | +## What It Does |
| 6 | + |
| 7 | +Point the **Tech Lead Orchestrator** at any feature request and it: |
| 8 | + |
| 9 | +1. **Detects your stack** — inspects `package.json`, `requirements.txt`, `go.mod`, and other config files via the Project Analyst agent. |
| 10 | +2. **Produces an Agent Routing Map** — an explicit, prioritized list of which specialists to invoke and in what order. |
| 11 | +3. **Presents findings for approval** — the human-in-the-loop gate prevents destructive actions without sign-off. |
| 12 | +4. **Delegates to specialists** — each specialist returns structured findings that the main agent coordinates into a complete solution. |
| 13 | + |
| 14 | +## Agents Included |
| 15 | + |
| 16 | +| Layer | Count | Examples | |
| 17 | +|---|---|---| |
| 18 | +| Orchestrators | 3 | tech-lead-orchestrator, project-analyst, team-configurator | |
| 19 | +| Framework Specialists | 13 | laravel-backend-expert, django-api-developer, rails-activerecord-expert, react-nextjs-expert, vue-nuxt-expert | |
| 20 | +| Universal Experts | 4 | backend-developer, frontend-developer, api-architect, tailwind-css-expert | |
| 21 | +| Core Quality | 4 | code-archaeologist, code-reviewer, performance-optimizer, documentation-specialist | |
| 22 | + |
| 23 | +## Quick Start |
| 24 | + |
| 25 | +```bash |
| 26 | +# 1. Clone and link agents into your Claude Code installation |
| 27 | +git clone https://github.com/vijaythecoder/awesome-claude-agents.git |
| 28 | +mkdir -p ~/.claude/agents |
| 29 | +ln -sf "$(pwd)/awesome-claude-agents/agents/" ~/.claude/agents/awesome-claude-agents |
| 30 | + |
| 31 | +# 2. Configure your project |
| 32 | +claude "use @agent-team-configurator and optimize my project to best use the available subagents." |
| 33 | + |
| 34 | +# 3. Build a feature |
| 35 | +claude "use @agent-tech-lead-orchestrator and build a user authentication system" |
| 36 | +``` |
| 37 | + |
| 38 | +## Key Capabilities |
| 39 | + |
| 40 | +- **Auto stack detection** — no manual configuration required |
| 41 | +- **Framework-specific expertise** — Django ORM, Rails ActiveRecord, Next.js SSR, Vue Pinia, Laravel Eloquent |
| 42 | +- **Security-aware reviews** — severity-tagged (critical / high / medium / low) |
| 43 | +- **Human-in-the-loop** — approval gate before execution on complex changes |
| 44 | +- **Structured handoffs** — each agent outputs parseable Markdown so context flows cleanly |
| 45 | + |
| 46 | +## Token Usage Note |
| 47 | + |
| 48 | +Multi-agent orchestration is token-intensive. Complex features may consume 10–50k tokens per workflow. Monitor your Claude subscription usage accordingly. |
| 49 | + |
| 50 | +## License |
| 51 | + |
| 52 | +MIT — use freely in your projects. |
0 commit comments