A 5-step workflow for using Claude to build features
Claude is used through a 5-step process for every new feature. A rough idea first becomes a written spec with no unclear missing parts. The plan is reviewed before any code is written, with extra review passes when the feature is risky.
During the , only one task is implemented at a time, and each task must be reviewed, tested, and checked in the real running app before the next task starts. The finished feature then goes through QA in the real app and only passes if there are no major bugs. The final step opens a PR, runs final , and records what was learned.
A separate cycle log records decisions the AI made on its own, so the work can be checked quickly without rereading every message. If any step loops 3 times without ing a result, the process stops and asks for human help instead of running forever.
Key points
- Each feature moves through spec, plan review, , QA, and ship steps.
- The allows only one task at a time before review and .
- Work must be checked against the real running app, not just assumed correct.
- A cycle log captures AI-made decisions for quick review later.
- If the same step fails to converge after 3 tries, the process stops and asks a human.