we collected all this data into a benchmark and built a free grader to let you know where you stand.

you answer on a 1–5 scale (e.g., autonomy runs from "suggestions only" to "agents own multi-hour workflows across code, infra, and external systems") - takes about 5 minutes.

https://agent-benchmarks.com/software-factory/

waiting for your results!