HUMANLY: A Configurable and Traceable Environment for Human-AI Collaborative Writing
Humanly Community is the open-source, self-hostable distribution of Humanly, a writing provenance platform for human-AI collaboration. Writers draft inside an owner-configured environment while Humanly records the process as it happens. The resulting certificate lets readers inspect how the document was produced instead of inferring authorship from the final text alone.
Visit the GitHubProduct
Humanly has two first-party web apps and a public certificate surface:
- Writer Portal: app.writehumanly.net for personal writing, assigned tasks, PDF-backed writing, AI-assisted drafting, submissions, and certificates.
- Publisher Portal: admin.writehumanly.net for creating tasks, configuring writing environments, distributing share links, reviewing submissions, and inspecting certificates.
- Certificate verification: shareable certificate URLs expose only the display fields selected by the certificate owner and verify the integrity seal attached to the record.
Humanly Community contains the complete Writer, Publisher, and Verifier workflow plus the tools needed to run it on your own infrastructure.
Features
- Configurable writing environments for personal writing and assigned tasks, including AI policy, copy-paste rules, timing, length bounds, writing instructions, PDF access, detectors, and assigned-task attempt rules.
- AI policy controls that can disable AI, allow only selected-text polish, allow only chat, or allow full in-platform assistance.
- Process tracing for typing, editing, copy/paste, focus, navigation, workspace activity, and in-platform AI interactions.
- Task distribution through invite codes and public share links, including account or guest participation depending on the task setting.
- Shareable certificates with authorship statistics, replay, environment settings, anomaly behavior review signals, and integrity seal details.
Humanly: A Configurable and Traceable Environment for Human-AI Collaborative Writing
Shenzhe Zhu, Haoqian Zhang, Xu Yang, Jingyu Tang, Yi Nian, Xiaoxue Du, Shu Yang, Alex Pentland, Joachim Baumann, Jiaxin Pei