joshft/correctless — The agent that writes the code never reviews it.

Description and use

**correctless** is an open‑source framework that enforces spec‑driven test‑first development for AI‑generated code, separating the code‑writing agent from a dedicated adversarial QA agent that reviews and challenges the output. By applying dynamic rigor and continuous, automated testing, it ensures that code produced by Claude (or similar models) meets security and correctness standards without relying on human post‑review.

Problem

joshft/correctless helps remove repetitive manual operations from a workflow.

Outcome

Early or unclear: treat as research material until maintenance, releases, docs, and issue activity are verified.

Workflow

• remove manual work
• connect tools into repeatable flows
• schedule operational tasks

Evidence

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

Source