Give your coding agent the power to write and run agent evals.
raindrop-ai/workshop is a local debugging tool that records an AI coding agent’s tokens, tool calls, and other execution spans as live traces. Developers use it to inspect agent behavior, create evaluations against their codebases, replay production traces, and iterate until tests pass. Its catalogue skills and instruction connect coding agents to Workshop’s tracing and evaluation workflows.
This repository also configures its own agents. See what workshop tells them →
Skill Claude CodeCodex
Set up Raindrop AI traces for an agent and verify they flow into Workshop. Use when the user says "set up traces", "add tracing", "instrument my agent", "wire Raindrop", "Workshop is empty", or "make traces show up in Workshop." Guides unknown repos by discovering runtime and telemetry setup, using current Raindrop…
Skill Claude CodeCodex
Set up a local agent replay server for Raindrop Workshop. Use when the user wants Workshop to replay a captured trace against their real local agent code and tools. Creates/updates .raindrop/agents.yaml, scaffolds a language-appropriate replay server, registers the project with raindrop replay register, and verifies…