ThinkFlowLab/system1-agents ? reverse-engineered prompt
Reverse engineered prompt
Build me a Python project that gives agents a fast, simple decision brain for everyday tasks like browser use, computer use, games, and basic robotics style actions. I want to be able to give it a task in plain English, have it pick the best next action, and either run a ready made agent or let me build a new one for a specific job.
Please include a command line way to make one decision, run a few example agents, and compare them against a normal chat model. It should also work as an MCP server, and be easy to plug into Claude Code or Codex so the host can hand off simple choices to this system. Add a few working examples like web browsing, desktop control, ticket routing, and a couple of games, plus clear docs on how to set it up and use it. If you need to look up current docs online, go ahead.
Are you gonna build this?
make sure you review the code using coderabbit