Agent Skills

Generate and edit images using Google's Nano Banana 2 (Gemini 3.1 Flash Image Preview) API. This skill should be used when the user asks to create or modify images, especially when they need fast iteration, explicit aspect-ratio control, or resolution control from 512px to 4K.

Install

npx skills add https://github.com/intellectronica/agent-skills --skill nano-banana-2
SKILL.md

Nano Banana 2 Image Generation & Editing

Generate new images or edit existing images with Nano Banana 2 (gemini-3.1-flash-image-preview).

Usage

Run the script using absolute or workspace-relative path (do NOT cd into the skill directory first):

Generate new image:

uv run ./.agents/skills/nano-banana-2/scripts/generate_image.py --prompt "your image description" --filename "output-name.png" [--resolution 512px|1K|2K|4K] [--aspect-ratio RATIO] [--api-key KEY]

Edit image(s) with references:

uv run ./.agents/skills/nano-banana-2/scripts/generate_image.py --prompt "editing instructions" --filename "output-name.png" --input-image "path/to/input1.png" [--input-image "path/to/input2.png" ...] [--resolution 512px|1K|2K|4K] [--aspect-ratio RATIO] [--api-key KEY]

Always run from the user's current working directory so outputs are saved in the active project context.

Model

Use model code:

  • gemini-3.1-flash-image-preview (Nano Banana 2)

Do not use gemini-3-pro-image-preview in this skill.

Resolution Options

Nano Banana 2 supports:

  • 512px (fastest, low-latency iteration)
  • 1K (default)
  • 2K
  • 4K

Map common user language:

  • "quick draft", "thumbnail", "tiny", "512", "0.5K" → 512px
  • no resolution mention → 1K
  • "2K", "medium", "normal" → 2K
  • "4K", "high-res", "ultra", "print quality" → 4K

Use uppercase K for 1K, 2K, 4K.

Aspect Ratios

Supported aspect ratios:

  • 1:1, 1:4, 1:8, 2:3, 3:2, 3:4, 4:1, 4:3, 4:5, 5:4, 8:1, 9:16, 16:9, 21:9

Map common requests:

  • square post / icon → 1:1
  • phone wallpaper / vertical reel → 9:16
  • widescreen / slide / hero image → 16:9
  • cinematic wide → 21:9
  • portrait print → 2:3 or 3:4
  • panorama banner → 4:1 or 8:1

If unspecified, let the model default behavior apply.

Reference Images (New in Gemini 3 Image workflows)

Provide up to 14 reference images when needed by repeating --input-image:

  • preserve character consistency
  • preserve object fidelity
  • combine multiple visual references into one output

Use one image for simple edits; use multiple images for composition or consistency-sensitive tasks.

API Key

Resolve API key in this order:

  1. --api-key argument
  2. GEMINI_API_KEY environment variable

If no key is available, stop and report a clear error.

Filename Generation

Generate filenames as: yyyy-mm-dd-hh-mm-ss-name.png

Examples:

  • 2026-02-26-17-31-04-japanese-garden.png
  • 2026-02-26-17-31-59-social-banner.png

Prompt Handling

For generation, pass the user's request as-is unless critically underspecified. For editing, include explicit transformation instructions in the prompt and preserve the original intent.

Output

  • Save PNG to the current directory (or to a provided path in --filename)
  • Return the full saved path(s)
  • Do not read the output image back unless explicitly requested

Examples

Create a 4K widescreen image:

uv run ./.agents/skills/nano-banana-2/scripts/generate_image.py --prompt "Futuristic city skyline at blue hour with neon reflections" --filename "2026-02-26-17-45-00-futuristic-skyline.png" --resolution 4K --aspect-ratio 16:9

Edit with multiple references:

uv run ./.agents/skills/nano-banana-2/scripts/generate_image.py --prompt "Create a polished campaign image that keeps the exact logo details and character identity" --filename "2026-02-26-17-50-10-campaign-image.png" --input-image "logo.png" --input-image "character.png" --resolution 2K --aspect-ratio 4:5

Related skills

ai-image-generationgenmedia-labs713KGenerate and edit images on RunComfy via the `runcomfy` CLI — a smart router across the full image-model catalog: FLUX 2 (Klein 9B/4B, Pro, Dev, Flash, Turbo, Max), Google Nano Banana 2 / Pro, OpenAI GPT Image 2, ByteDance Seedream 5 / 4-5 / 4-0 and Dreamina 4-0, Alibaba Qwen Image and Z-Image Turbo, Wan 2-7. Covers both text-to-image (t2i) and image-to-image / edit (i2i) endpoints — the skill picks the right model for the user's actual intent (typography precision, photoreal portraits, sub-secoai-image-generation101-skills547KGenerate AI images with GPT-Image-2, FLUX, Gemini, Grok, Seedream, Reve and 50+ models via inference.sh CLI. Models: GPT-Image-2, FLUX Dev LoRA, FLUX.2 Klein LoRA, Gemini 3 Pro Image, Grok Imagine, Seedream 4.5, Reve, ImagineArt. Capabilities: text-to-image, image-to-image, inpainting, LoRA, image editing, upscaling, text rendering. Use for: AI art, product mockups, concept art, social media graphics, marketing visuals, illustrations. Triggers: flux, image generation, ai image, text to image, stnano-banana-2prime-skills424KGenerate images with Google Nano Banana 2 (Gemini-family flash-tier text-to-image) on RunComfy — bundled with the model's documented prompting patterns so the skill gets sharper output than naive prompting against the same model. Documents Nano Banana 2's strengths (rapid iteration, in-image typography rendering, predictable framing, optional web-grounded context), the resolution-tier pricing, the safety-tolerance dial, and when to route to Nano Banana Pro / GPT Image 2 / Flux 2 / Seedream insteimage-editprime-skills424KEdit images on RunComfy — this skill is a smart router that matches the user's intent to the right edit model in the RunComfy catalog. Picks Nano Banana Edit (batch up to 20, identity-preserving default), OpenAI GPT Image 2 Edit (multilingual in-image text rewrite, multi-ref composition, layout precision), Flux Kontext Pro (single-ref high-fidelity local edit), or Z-Image Turbo Inpaint (mask-driven precise region edit). Bundles each model's documented prompting patterns so the skill gets sharper

Search skills and MCP servers

Fuzzy search across 23,137 skills and servers