Testing a two-model workflow in Claude Code
A Claude Code is being tested with two separate models. The first model takes a rough user request and turns it into a clearer, more complete . The second model receives that improved and starts building the code.
The main idea is to split the work into a ‘request cleanup’ step and a ‘code building’ step, so vague instructions cause fewer mistakes. No results, setup details, or measured benefits are included yet; it is still an early experiment looking for experience from others.
Key points
- The uses two models inside a Claude Code setup.
- The first model rewrites a rough request into a stronger .
- The second model uses that to build the code.
- The goal is to reduce mistakes caused by unclear instructions.
- The idea is still unproven in the shared item, with no concrete results yet.