My core axioms:
1. Solving pre-specified problems will become much less valuable as AI rapidly improves and becomes unfathomably intelligent.
2. Humans will continue to play an important role in steering problem solving, especially through deciding what is worth pursuing and exploring a problem space.
3. We have an incomplete and inadequate understanding of how great scientists choose what to work on and how they explore, or play with, a problem space.
See: https://t.co/49O8TWAmaz
To install: run npx engelbart-cli in your CLI, restart claude, then run /goals-ui in an existing Claude Code chat.
(10/10)
Goals and intent are usually implicit when you work with coding agents, causing information to be jumbled as projects grow in size.
We built Engelbart, an open-source tool that gives you and Claude Code a shared, editable representation of what you’re trying to build. (1/10)