关键词: OpenClaw, 开源协作, 贡献结构

Abstract:

AI-native (Artificial Intelligence-native) open-source projects are evolving beyond code repositories into capability ecosystems that integrate models, tools, plugins, and user scenarios. However, quantitative evidence on whether an early surge in attention translates into stable collaborative governance capacity is limited. Using publicly available OpenDigger data, this study examines OpenClaw’s early collaboration dynamics from November 2025 to March 2026 across four dimensions: project influence, participation scale, issue-handling capacity, and contribution distribution. These dimensions are then compared with those of three AI-native open-source projects—vLLM, Claude Code, and Dify—over the same observation window. The results show that OpenClaw’s growth in attention translated into substantive collaboration expansion. In the first quarter of 2026, both its OpenRank and activity surpassed those of the three open-source projects. However, OpenClaw’s issue close rate was 0.70—the lowest among the four projects—indicating that its issue-handling capacity could not keep up with the growth in participants and collaboration events. At the organizational level, activity was highly concentrated in the core repository, while contributions within that repository exhibited a long-tail distribution. A relatively high bus factor suggests broad contributor participation; however, its governance implications must be assessed in conjunction with the structure of the core maintainer group. These findings suggest that, as AI-native open-source projects expand participation, they need to develop corresponding mechanisms for issue handling and contribution absorption.

Key words: OpenClaw, open-source collaboration, contribution structure