# CaP-X: Benchmarking and Improvement Framework for Robot Manipulation Coding Agents Released

- Source: Berkeley AUTOLAB
- Published: 2026-03-23T18:08:10.000Z
- RoboHot: https://robohot.news/en/items/ij09i2e21o9bn9yq5isel1ycd
- Original: https://arxiv.org/abs/2603.22435

## Summary

Teams including Berkeley AUTOLAB propose CaP-X, an open framework for systematically studying Code-as-Policy agents in robot manipulation, centered on the interactive environment CaP-Gym and CaP-Bench, which evaluates 12 frontier models.

## Why it matters

The original text presents a systematic evaluation framework for Code-as-Policy agents and the finding that test-time scaling can mitigate abstraction dependence, offering reference value for research on coding-agent manipulation control.