Media
2 items
Description
See which lines in any GitHub repo are hardcoded guesses that should be a Jev judgment, with a fit score for each. Read-only. JevX finds the judgment calls hiding as rules in code: regexes that guess what an error means, keyword lists that guess what a customer wants, includes() checks that guess intent. Open any GitHub repository, click JevX, and get: The exact file and lines where Jev fits The current code, and the question Jev would ask instead A fit score from your AI, patterns from real Jev projects, and TypeSafe (optional) A plain verdict: "Jev fits here", "could help", "mixed signals" or "not needed" A ready prompt for Claude or a JevX command to apply the change locally, with your tests running before and after Bring your own AI key: Gemini, xAI (Grok) or OpenRouter. Private by design: read-only, keys stay in your browser, code goes only to the AI provider you choose, and secrets are removed first. No tracking, no analytics. Open source (MIT): https://github.com/vij-sameerb5/JevX Website: https://jevx.live
rating_count is the Chrome Web Store ratings count, not a written-review count.