OpenHands

OpenHands added Agent Client Protocol support to its Agent Canvas, SDK, and Cloud, letting teams run different coding agents through one interface across local, remote, and cloud backends. The release also underpins new OpenHands Index results, so teams can compare harness-plus-model combinations instead of model-only runs.

Hermes Agent added an OpenHands orchestration skill that can be installed with a single command. The addition matters because Hermes can now route work across OpenHands, Claude Code, Codex, and OpenCode through the same skills interface.

OpenHands introduced EvoClaw, a benchmark that reconstructs milestone DAGs from repo history to test continuous software evolution instead of isolated tasks. The first results show agents can clear single tasks yet still collapse under regressions and technical debt over longer runs.

OpenHands published a skill-eval recipe with bounded tasks, deterministic verifiers, and no-skill baselines, then showed some skills speed agents up while others make them brittle. Teams shipping skill libraries should measure them per task and model before rollout.