Wunderlabs

We build tools that agents use to ship reliable systems.

Harnesses

Test agent behavior across real workflows, regressions, and long-running tasks with reproducible harnesses.

Agentic Dev Envs

Run and modify generated codebases in isolated, deterministic environments with zero local setup.

Human-in-the-Loop

Inspect traces, review tool calls, replay failures, and keep humans involved where reliability matters.

Projects

Tools for reliable agents.

Open Source
All systems operational

Claudebin

Publish Claude Code sessions with a single command and get a shareable link with syntax highlighting, tool calls, and the full conversation thread.

Open Source
All systems operational

Openable

Import, run, and edit full generated codebases without installing Node.js, configuring environments, or using the terminal.

Community

A community for agentic builders.

Through agentic.tm

200+

engineers, researchers and academics working on agent-driven development

Let’s build together

Open source by default. Ship with us.

Hiring agentic native engineers and researchers focused on infrastructure, reliability, and agent systems.

Based in Europe
Software lab