GitHub - raindrop-ai/workshop: Give your coding agent the power to write and run agent evals.

The local debugger your agent is missing. Watch your agent think locally, the moment it happens: every token, every tool call, every decision.

Give Claude Code the power to read your traces, write evals against your codebase, and fix what's broken.

Install

curl -fsSL https://raindrop.sh/install | bash

Build from source

git clone https://github.com/raindrop-ai/workshop.git
cd workshop
bun install
bun run dev

bun run dev starts the local Workshop daemon and Vite UI. Open http://localhost:5899 after it starts.

Instrument your agent

Open your coding agent of choice in your repository and run:

This will instrument your agent with Raindrop tracing and open Workshop in your browser.

That's it. Traces stream into the UI the moment your agent runs.

What it does

• Live streamed traces. Every token, tool call, and span streams into Workshop as it happens. No polling, no refreshing.
• Coding-agent integration. Claude Code reads your traces, writes evals against your codebase, and fixes what's broken.
• Self-healing eval loop. Claude writes the eval, runs your agent, sees the failure, fixes the code, and re-runs — until every assertion passes.
• Local replay. /setup-agent-replay scaffolds an HTTP endpoint that replays a production trace against your real agent code.

Compatible with everything

• Languages: TypeScript, Python, Go, Rust
• SDKs: Vercel AI SDK, OpenAI Agents SDK, Anthropic SDK, Claude Agent SDK, LangChain, LangGraph, CrewAI, Mastra, Pydantic AI, DSPy, Google ADK, Strands, Agno, Deep Agents
• Providers: AWS Bedrock, Azure OpenAI, Vertex AI
• Coding agents: Claude Code, Codex, Devin, Cursor, OpenCode

Configuration

CLI

raindrop workshop          # start and open UI
raindrop workshop setup    # write .env, then start and open
raindrop workshop status   # check health
raindrop workshop reset    # delete local DB after confirmation
raindrop update            # update the binary

License

MIT.