Call a Breaking Change a Patch, and Claude Sonnet Checked It Only 2 Times in 10

Listen through the publisher, then discuss the episode here or arrange an in-person Roundtable. Your city determines where to meet.
Roundtables are designed for groups of four, with a shared subject and room for everyone to speak. Come for discussion, with no dating expectation.
Microsoft gave AI coding agents the same breaking library upgrade, labelled either 1.4.3 or 2.0.0. Claude Sonnet 5.5 checked the major version in all 10 runs but the patch in only 2, often leaving the app broken, while Claude Opus 5.5 checked every time. Hear what 540 review-and-merge runs showed, why four merges broke main under patch or minor labels, how to share Claude Code settings with your team (shared vs local files, and why ask rules apply before folder trust), and a simple rule so version labels never decide how much testing happens. Reporting checked 13 November 2026 from Microsoft's published experiment and Anthropic's Claude Code settings documentation. The test used GitHub Copilot CLI, not Claude Code; the suggested rules are recommendations, not tested products. Includes a Railway affiliate advertisement. Sources: https://developer.microsoft.com/blog/do-ai-coding-agents-care-about-semver/ https://code.claude.com/docs/en/settings Includes a Railway affiliate advertisement. Find our links at https://aishows.cc; we may earn a commission.
Episode information is provided by the publisher. Listening takes place on the publisher’s service.
Listen at the source ↗









