Recently, Peter Steinberger, the developer of the popular AI project OpenClaw, publicly accused Tencent on the social platform X of scraping all the skill data from his ClawHub platform without authorization, to build its own "SkillHub" platform. Steinberger stated that Tencent made a full copy of the original project without providing any substantial support, and even some personnel had emailed to complain about the "access rate limits" of ClawHub, which hindered their scraping efficiency.
In response to these allegations, the official Tencent AI account quickly clarified, emphasizing that the initial purpose of SkillHub was to create a localized skill mirror platform based on the OpenClaw ecosystem, aiming to solve availability issues such as access latency for Chinese users. Tencent disclosed technical data showing that SkillHub handled 180GB of traffic (approximately 870,000 downloads) in its first week, while actually only pulling 1GB of data from the official source, objectively reducing the bandwidth pressure on the original site by 99.4%.
Additionally, Tencent emphasized that its team members have always been active contributors to the project and is willing to deepen the collaboration in the form of sponsorship in the future. Although some believe that the mirror site significantly reduces the operational costs of open-source projects, Steinberger insists that respecting open-source licenses and developers' rights to be informed is a prerequisite for cooperation. Both sides should reach an official certification through synchronized collaboration, rather than one-sided mirroring.
This dispute reflects the conflict between developers' original rights and the expansion of big companies' ecosystems during the boom period of large models and AI agents. While pursuing the efficiency of technological implementation, how to establish a more "polite" and transparent open-source collaboration mechanism will remain a continuous topic of exploration for the AI industry.
