Blackbox records each model call and tool call your agent makes. When a run goes wrong, rewind to the exact step, change one thing, and fly it again.
Bright orange so you can find it after a crash. Works with Anthropic and OpenAI message logs and OpenTelemetry traces.
The same channels for every run, so you can line two flights up and see exactly where they split.
| Channel | Unit | Why you want it |
|---|---|---|
| Model calls | ms, stop reason | What the model said, how long it thought, and whether it finished or hit a limit. |
| Tool calls | name, input, output | Every argument and every result, including the errors the agent quietly ignored. |
| Tokens | in / out | Where the context ballooned, and which step made the run expensive. |
| Cost | USD, estimated | Per step and per flight, at the prices you set. |
| Flags | loop, error, slow, claim | Repeated identical calls, error streaks, slow tools, cut-off answers, and actions the model claims without a tool to back them. |
Pick any step and fork. Edit the prompt, a tool result or the model, and Blackbox replays the run from there as a new branch you can compare side by side.
Built-in detectors catch the obvious failures. Claude reads the whole tape for the rest and writes it up like an investigator.