joshft/correctless

61 stars · Last commit 2026-05-31

The agent that writes the code never reviews it. Spec-driven TDD with agent separation, adversarial QA, and dynamic rigor for Claude Code.

README preview

<p align="center">
  <img src="assets/logo.png" alt="Correctless" width="400">
</p>

[![OpenSSF Scorecard](https://api.scorecard.dev/projects/github.com/joshft/correctless/badge)](https://scorecard.dev/viewer/?uri=github.com/joshft/correctless)
[![CI](https://github.com/joshft/correctless/actions/workflows/ci.yml/badge.svg)](https://github.com/joshft/correctless/actions/workflows/ci.yml)
[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
[![Skills: 32](https://img.shields.io/badge/skills-32-blue.svg)](docs/skills/)
[![Version: 3.0.0](https://img.shields.io/badge/version-3.0.0-green.svg)](CHANGELOG.md)

Composable [Claude Code](https://docs.anthropic.com/en/docs/claude-code) skills that enforce a correctness-oriented development workflow. Spec before you code. Test before you implement. Never let an agent grade its own work.

Built with Correctless.

## The Problem

AI coding assistants are fast but sloppy. They write code that works for the happy path, skip edge cases, and silently introduce bugs that don't surface until production. The same model that wrote the code will review it and say "looks good" — because it's confirming its own decisions.

Correctless fixes this by structuring the workflow so that **every phase is executed by a different agent with a different lens**:

View full repository on GitHub →