New: Quality Governance, a quality gate for the code your AI agents write. Complexity, function by function. Every code scan now measures cognitive and cyclomatic complexity, copied blocks (100 tokens over 10 lines or more) and test coverage from your own CI. JavaScript, TypeScript and Python. A gate on every pull request: leaked secrets, vulnerable or malicious dependencies, security issues, and limits on duplication and complexity in new code. Pick Relaxed, Recommended or Strict. Report only, or block the merge on GitHub until it passes. On every plan, including the Free Week: 1 pull request a day on Free and Watch, every one on Guard and Shield. Watch the video: bit.ly/vd-quality-gate-video-trustmrr Try it: bit.ly/vd-quality-gate-trustmrr

What we shipped in October: Find us in Claude’s connector directory. VibeDoctor is listed at claude.ai/directory/vibedoctor. Add it, sign in once, and Claude can check your code, show open findings and explore your code map. For Claude Code, Codex and Cursor setup, see vibedoctor.io/connectors. Connect your AI agent with one paste. Each project gets its own setup line. Hooks for Claude Code, Codex, Cursor, VS Code and Devin give your agent the project summary when a session starts, and a file’s open findings the first time it edits that file. Find in code. Search your app in plain words and get the file and line, plus what calls it and what it calls. A new Home. Scores, your latest push, X-Ray, Vibe Story, speed and monitoring on one screen. Security badges. Earn badges as your project passes checks, add one to your site, and share a public check page. Full Lighthouse report. Every audit on mobile and desktop, passed ones included, with Agentic Browsing and earlier runs.