Glad you're here. ImageAI gets better every time someone reports a bug, suggests a feature, or sends a pull request — and all three are welcome, whether it's your first contribution or your fiftieth.
The fastest way to kick an idea around is the Chameleon Labs Discord. That's where we talk features, share AI art and help each other with setup. No formality needed — "wouldn't it be cool if..." is a perfectly good opener. If the idea has legs, we'll help you shape it into an issue (or you can just build it and open a PR).
Here's a perk worth knowing about: when you file a detailed issue or open a pull request, our automated tooling analyzes it against the actual codebase. It checks whether the bug was already fixed, traces the likely cause and often posts a diagnosis or review without you having to wait for a human.
The more you give it, the more you get back. Steps to reproduce, log output and version info produce a real analysis. "It doesn't work" produces a shrug.
Before filing:
- Search existing issues — someone may have beaten you to it.
- Update to the latest version and confirm the bug is still there.
- Grab your environment details — OS, ImageAI version, Python version and which provider you were using (Google, OpenAI, Stability or Local SD).
Handy tip: ImageAI copies the most recent session log to ./imageai_current.log in your working directory when it exits. Attach it. It usually tells us exactly what went wrong.
Short and specific. [Bug] File upload fails with special characters beats Help, not working! every time.
Summary: one line describing the issue.
Environment: OS, ImageAI version, Python version, provider.
Steps to reproduce: numbered steps, starting from launch if needed.
Expected vs. actual: what you thought would happen, and what happened instead (include crashes, error text or log excerpts).
Reproducibility: always, intermittent or rare.
Attachments: screenshots, imageai_current.log, or a minimal example that triggers the bug. Small reproductions get fixed fastest.
Feature ideas are welcome as issues too — but for anything fuzzy or early-stage, Discord is the better first stop. A quick conversation there often saves a lot of back-and-forth in the issue thread. When you do file a feature issue, describe the problem you're trying to solve, not just the solution you have in mind. (Sometimes there's already a way to do it, and sometimes the real fix is even better than the first idea.)
Want to fix it yourself? Even better.
- Fork the repo and create a branch.
- Set up your environment:
python -m venv .venv, activate it, thenpip install -r requirements.txt(details in the README). - Keep the PR focused — one fix or one feature per PR.
- Describe what changed and why. Remember, detailed PRs get that same automatic analysis, so a clear description means a faster, better review.
Not sure whether a change would be accepted? Ask first — on Discord or in an issue. Nobody wants you to spend a weekend on something that needed a five-minute conversation.
Keep an eye on the thread and answer follow-up questions when they come. If your issue gets resolved, a quick note about what fixed it helps the next person who hits the same thing.
Thanks for helping make ImageAI better. See you on Discord!