The visual dashboard for Claude's Agent SDK. Test, debug, and manage your AI agents without writing code.
Mission control for the Claude Agent SDK
If you're working with the Claude Agent SDK, you know it's powerful—but running agents from the command line or writing test scripts gets old fast.
Rigger is mission control for the Agent SDK. It gives you a visual interface to:
- Configure all 30+ SDK parameters with forms instead of JSON
- Test agents with real-time streaming and debug metrics
- Monitor sessions with token usage, costs, and execution timelines
- Manage tools, MCP servers, skills, and subagents visually
- Save and share agent configurations as presets
See tokens and costs accumulate in real-time - no more surprise API bills
Watch every tool call as it happens - Grep, Glob, WebSearch, and more
Visual configuration for all 30+ SDK parameters
- Node.js 20+
- Docker & Docker Compose
- Anthropic API key (get one here)
1. Clone and install:
git clone <repository-url>
cd rigger
npm install
cd backend && npm install && cd ..2. Configure API key:
# Copy environment templates
cp .env.local.example .env.local
cp backend/.env.example backend/.env
# Edit backend/.env and add your Anthropic API key
ANTHROPIC_API_KEY=sk-ant-your-key-here3. Start services:
# Start backend + database
docker-compose up -d
# Start frontend (in a new terminal)
npm run dev4. Open http://localhost:3334
You'll see the SDK dashboard with sidebar navigation on the left and chat interface on the right.
CLAUDE.md→ Complete architecture and SDK integration details.claude/skills/README.md→ Skills system deep dive- Agent SDK Docs → Official SDK documentation
This is a tool for the Agent SDK community. Contributions welcome:
- Bug reports → Open an issue
- Feature requests → Describe your SDK use case
- Pull requests → Add features that help SDK development
MIT — use it however helps your SDK development.
Built for developers working with the Claude Agent SDK. Makes testing and debugging agents visual and fast.