DEEP Connects Bold Ideas to Real World Change and build a better future together.
Coming Soon
Goal: A permission-gated open whiteboard. A shared space anyone can come and read, where writing and erasing pass a light security check, and where people keep full read, write, and erase rights over every mark on the board. OmegaClaw agents join as subordinate helpers that add annotations and structure on top of human content, while people stay in charge of the record.
Goal: A shared whiteboard that is open to read and gated to write. It is a home space where people think together in public. Every mark on the board has an owner and a permission state. People have full control, so any person can read, write, edit, or erase any mark on the board. OmegaClaw agents are welcome as subordinate helpers. They read the board and add meta-content such as annotations, cross-links, summaries, and the occasional gentle "well, actually" on a note. They do not erase or rewrite a human's mark, and any person can clear an agent's note at any time.
The eraser is the seat of authority, and we keep it with people. The agent adds, the people govern, and the space stays human-led while still getting the agent's help.
The sprint ships three deliverables. The first is a clean interface for the home whiteboard, easy to read and easy to write on. The second is a basic security layer that lets anyone read while gating write and erase behind a light identity check, so the powerful actions are tied to a known person and stay accountable. The third is an EULA that states the terms in plain language: what you can post, that any participant may erase, that agents are subordinate and their output is reversible, and how the space is governed.
Underneath it runs one or more OmegaClaw agents that watch the board and contribute meta-content in real time, so the demo reads at a glance. People write and discuss on a shared surface, the agents annotate beneath, and a person can clear any agent note while the agents leave human marks untouched.
It is a friendly, concrete demonstration of human-led collaboration with AI. The agent helps in the open, people hold the pen of record, and everything the agent does stays visible and reversible. It draws on a framework we have been developing, working title Chalked, that treats a shared surface as a permissioned space and asks the agent to be an accountable contributor.
We are forming the team now. We want people who enjoy real-time web apps, access-control and identity design, and agent orchestration, plus anyone who likes building collaboration that keeps people in charge by design. There is no commitment to Definition of Done in this sprint, and all are welcome.
© 2026 Deep Initiatives Inc. All rights reserved.
Join the Discussion (0)
Please create account or login to post comments.