NanoBananaGemini
AGENTS.md0 starsChanged 28 days agoMITView on GitHub
This repository owns the Nano Banana Gemini CLI manifest, native sensitive Token setting, workflow skill, commands and bilingual onboarding. The existing NanoBananaMCP owns the server; reuse its hosted endpoint without duplicating/changing shared code. Do not add hooks, trust bypass, hidden fallbacks or automatic paid generation. Run python3 -m unittest discover -s tests, python3 scripts/package_extension.py --output /tmp/nano-banana-by-ace-data-cloud.tar.gz, the official gemini extensions validate and isolated lifecycle smoke checks. Ordinary CI must use test discovery. Keep English/Chinese guidance aligned, preserve real screenshot sources…
- Reads credentials
In this repositoryOne other file it gives its agents.
- nano-banana-imageSkill
The file
Files
NanoBananaGemini
This repository owns the Nano Banana Gemini CLI manifest, native sensitive Token setting, workflow skill, commands and bilingual onboarding. The existing NanoBananaMCP owns the server; reuse its hosted endpoint without duplicating/changing shared code. Do not add hooks, trust bypass, hidden fallbacks or automatic paid generation.
Run python3 -m unittest discover -s tests, python3 scripts/package_extension.py --output /tmp/nano-banana-by-ace-data-cloud.tar.gz, the official gemini extensions validate and isolated lifecycle smoke checks. Ordinary CI must use test discovery. Keep English/Chinese guidance aligned, preserve real screenshot sources and license, do not commit .env, keys or private user transcripts. Publish changes through the bot PR; do not merge automatically.
Generation uses prompt; image editing uses prompt plus image_urls as a real array. Keep aspect_ratio/resolution/count and model within the selected schema. The service has no static reference tool: first setup checks inspect loaded tool names/schema only, without invoking generation or sending a fabricated task ID. Query only an existing original task ID supplied/authorized by the user. For edits, omitting aspect_ratio preserves the first image aspect ratio; verify actual output dimensions rather than promising a resolution.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.