Skip to content

Latest commit

 

History

History
54 lines (49 loc) · 3.45 KB

priorities.md

File metadata and controls

54 lines (49 loc) · 3.45 KB

Project Priorities

This file outlines the current priorities for the Universal Constructor project.

Game plan

  • Get OpenAI API working
  • Create command line REPL that lets you talk to the assistant
  • Add basic ability to read and edit files so it can work on itself
  • Clean up, add a README and push to GitHub so others can try it
  • Make the search/edit functionality robust, so that UC can reliably commit changes
  • Simplify the tools code until UC can reliably add and edit its own tools
  • Add the ability to reliably verify if a given changeset is valid/working/good
  • Add the ability to make modest changes end-to-end (from description to tested, merged PR)
  • Add the ability to work on projects beside itself
  • Build the ability to run in a CI environment, so you can fire off a task that comes back with a PR
  • Build a meta testing environment that does CI on the CI tool, that tests which tasks it is up to
  • Improve UC until it can handle the easiest 20% of real tasks
  • Use UC to improve testing tools & UC itself until it can handle 50%
  • Make a new, small, real "UC-first" project build and maintained by UC with only high level direction.
  • Add ability to fully monitor projects in production
  • Add ability to make small fixes & self-deploy
  • Add ability to communicate directly with users and summarize information
  • Add ability to suggest its own tasks based on monitoring and user input
  • Start making small changes all the way from conception (based on monitoring & input)
  • Add ability to subjectively judge simplicity and quality, in addition to user feedback/monitoring
  • Start making small self-improvement fully autonomously (humans get a changelog)
  • Ability to complete most tasks a human software developer can
  • Ability to build, monitor, and iterate on most software that a human developer can
  • Ability to go from high level human objectives to working software & systems
  • Ability to, with human assistance, start and run a company
  • ... a profitable company
  • ... a company with IRL footprint & goals
  • ... a successful group of companies providing all reasonably required capital/income
  • ... full-stack company, able to bootstrap from raw materials to all physical needs
  • Achieve material abundance for a small group of humans. All "single player" needs/wants too cheap to meter.
  • Build a healthy, self-propagating, growing culture that inspires those humans to reach the stars
  • Expand abundance to a larger group of humans
  • Set up first permanent human habitation off-planet
  • Expand abundance to all humans who want it
  • Multiple permanent human habitations in the solar system that could survive the destruction of earth
  • First human interstellar voyage
  • TODO: what next?

Ideas for next small tasks:

  • Upgraded context awareness via system messages and/or simulated tool calls (in progress)
  • fix listening
  • Better tool call visibility
  • Clean up multiple messaging/pasting/the repl
  • Test (on e.g. fresh setup, other dirs) & merge to master
  • Better markdown behaviour with streaming responses
  • Get editFile to 95%+ reliable for small code changes (might be there; add challenges to test?)
  • Ability to work in a git branch, where small steps can get committed / rolled back, to allow more complex changes