image-nanobanana

name	image-nanobanana
description	Generate and edit production-ready, watermark-free image assets with Google's Nano Banana models — gemini-3.1-flash-image (Nano Banana 2) and gemini-3-pro-image (Nano Banana Pro) — via the paid Gemini API and the gemini-cli nanobanana extension. Use when the user wants to create, generate, make, draw, design, edit, restore, or iterate on any image, picture, photo, graphic, illustration, logo, or visual asset; mentions Nano Banana, Gemini image generation, text-to-image, or AI images; complains about the Gemini sparkle watermark or wants a watermark removed from an image; or asks for app icons, favicons, hero banners, OG/social cards, YouTube thumbnails, marketing creatives, diagrams, seamless patterns, sprite sheets, or consistent characters. Bundles preflight, a direct-API generator with aspect-ratio and 512/1K/2K/4K control, gemini-cli setup, prompting guidance, and per-asset production recipes.
license	MIT
compatibility	Bundled scripts need Python 3.8+ (standard library only), bash, and curl. A paid-tier Gemini API key is required for generation (image models have no free-tier quota). gemini-cli is optional — only the interactive extension path uses it. Linux, macOS, and WSL.
inputs	[{"name":"GEMINI_API_KEY","description":"Gemini API key from https://aistudio.google.com/apikey — its Google Cloud project must have billing enabled (Tier 1+); image models are not available on the API free tier. GOOGLE_API_KEY and NANOBANANA_API_KEY work as fallbacks. Instead of an env var, the key can be stored in ~/.gemini/.env via scripts/nb-cli-setup.sh --set-key (never in the skill folder — it is git-tracked).","required":true}]
metadata	{"author":"github.com/kryptobaseddev","version":"1.2.0","last_updated":"2026-06-11 19:00:00","category":"media"}
allowed-tools	Bash Read Write Edit Glob Grep

Nano Banana Image Asset Generation

Produce professional, production-ready image assets — icons, banners, thumbnails, OG cards, diagrams, patterns, character sets — with Google's Gemini image models ("Nano Banana"). Two execution paths, one skill:

1. Direct API via scripts/nb-generate.py — the production path: exact aspect ratios, 512/1K/2K/4K sizes, editing with reference images, retries, JSON output, cost estimates.
2. gemini-cli + nanobanana extension — the interactive path for users who live in a gemini-cli session (/generate, /edit, /icon, ...).

Both paths produce images with no visible watermark (the Gemini "sparkle" overlay exists only in the consumer Gemini app for free/Pro tiers). Every image carries Google's invisible SynthID provenance mark — leave it intact.

Facts that prevent broken work (June 2026)

Fact	Consequence
GA model ids: gemini-3.1-flash-image (Nano Banana 2), gemini-3-pro-image (Nano Banana Pro)	Use these. The -preview ids shut down 2026-06-25
The nanobanana extension still defaults to a -preview id	Always pin NANOBANANA_MODEL (nb-cli-setup.sh does it)
Image models have zero free-tier API quota	A key without billing gets 429s — that's a billing problem, not a bug
gemini-cli consumer OAuth retires 2026-06-18	API-key auth is the only durable gemini-cli setup
Visible sparkle watermark = consumer Gemini app (free/Pro) only	Never source production assets from the Gemini app; generate via API

Preflight (run once per session)

bash scripts/nb-preflight.sh          # human-readable; --quiet for JSON

It verifies python3/curl, finds the API key, probes the GA model with the key, and audits gemini-cli/extension/auth. Exit codes: 0 ready, 2 usage, 4 missing binary, 5 network, 6 key missing/rejected. One blind spot: preflight cannot prove billing — an un-billed key still passes and then 429s on the first real generation (that 429 means "enable billing", not "retry").

If there's no key: send the user to https://aistudio.google.com/apikey, remind them to enable billing on the key's project, then (it needs hidden keyboard input, so the agent must not run it):