Your coding agent isn't dumb. Your repo just never told it anything.
If Claude Code or Cursor keeps guessing the test command, breaking conventions, or asking the same setup questions, that's hours of review and rework every week. And it's rarely the model's fault.
I built WeaveAI Ready to find exactly where your repo leaves agents blind. Point it at a GitHub repo: 87 checks plus an agent pass in ~3 minutes, then it opens PRs that fix the gaps.
What's the one thing you wish your coding agent knew about your codebase?
#AI #ClaudeCode #DevTools #buildinpublic