DockSmith: Scaling Reliable Coding Environments via an Agentic Docker Builder
For software engineering agents, DockSmith addresses the bottleneck of reliable Docker environment construction, enabling scalable execution-grounded training and evaluation.
DockSmith introduces an agentic Docker builder that treats environment construction as a core capability, achieving 39.72% Fail-to-Pass and 58.28% Commit Rate on Multi-Docker-Eval, and improving performance on SWE-bench Verified, SWE-bench Multilingual, and Terminal-Bench 2.0.
Reliable Docker-based environment construction is a dominant bottleneck for scaling execution-grounded training and evaluation of software engineering agents. We introduce DockSmith, a specialized agentic Docker builder designed to address this challenge. DockSmith treats environment construction not only as a preprocessing step, but as a core agentic capability that exercises long-horizon tool use, dependency reasoning, and failure recovery, yielding supervision that transfers beyond Docker building itself. DockSmith is trained on large-scale, execution-grounded Docker-building trajectories produced by a SWE-Factory-style pipeline augmented with a loop-detection controller and a cross-task success memory. Training a 30B-A3B model on these trajectories achieves open-source state-of-the-art performance on Multi-Docker-Eval, with 39.72% Fail-to-Pass and 58.28% Commit Rate. Moreover, DockSmith improves out-of-distribution performance on SWE-bench Verified, SWE-bench Multilingual, and Terminal-Bench 2.0, demonstrating broader agentic benefits of environment construction.