Coding Agents Enable Automated And Manual Testing

This article explains that coding agents can execute and validate the code they generate, enabling iterative fixes and test-driven development. It outlines manual testing patterns (python -c, curl), browser automation with Playwright and agent-focused CLIs like Rodney, and documentation via Showboat's note/exec/image commands to capture commands, outputs, and screenshots. The piece recommends combining TDD, automated tests, and agentic manual testing to catch issues automated tests miss.