OpenClaw went viral. Meet the maintainers building and securing it.

OpenClaw, a personal AI assistant, rapidly transformed from a personal experiment into a viral, globally impactful open source project. Founded by Peter Steinberger as a weekend project in November 2025, the GitHub repository grew to an impressive 388,000 stars, 81,000 forks, and over 80,000 commits by August 26, 2026. In a recent video interview, Steinberger and several OpenClaw maintainers discussed the challenges and lessons surrounding managing a flood of pull requests, rethinking contributor trust and code review, addressing software supply chain risks, and balancing powerful agent capabilities with security.

Three key takeaways emerged from the conversation. First, the project experienced a significant shift from mere participation to high-quality contributions, as maintainers struggled to sift through thousands of pull requests and issues. Second, OpenClaw's maintainers sought to remain welcoming to new contributors, including non-developers and those using AI agents to help, by valuing promising ideas and refining contributors' work rather than dismissing imperfect submissions.

Lastly, agents proved beneficial in reclaiming time for maintainers, but they also introduced challenges in ensuring accountability and work-life balance.

Further lessons highlighted the importance of trust signals, with maintainers emphasizing the value of agent transcripts, screenshots, testing, and contributors' explanations as evidence of their understanding and consideration of the project's interactions. Additionally, OpenClaw maintainers began incorporating AI tools into their code review process, further streamlining the evaluation of AI-generated contributions.