agentleFS
bongartzdiazbongartzdiaz/agent-skills/skills/nano-banana-2/SKILL.md

nano-banana-2

Skill0 starsChanged 10 months agoCC0-1.0View on GitHub

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.

  • Reads credentials

9 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use flux-2-klein, flux-kontext and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • agentspace-soSkill

    flux-2-klein

    Generate images with Flux 2 Klein (Black Forest Labs' distilled fast variant of Flux 2) on RunComfy — bundled with the model's documented prompting patterns so the skill gets sharper output than naive prompting against the same model. Documents Flux 2 Klein's strengths (sub-second latency, multi-reference brand styling, declarative subject-first prompts), the step-count strategy (4–8 for fast iteration, ~25 for polish), the 9B vs 4B variant trade-off, and when to route to Flux 2 Pro / Seedream 5 / GPT Image 2 instead. Calls `runcomfy run blackforestlabs/flux-2-klein/9b/text-to-image` (or `/4b/`) through the local RunComfy CLI. Triggers on "flux 2 klein", "flux-2-klein", "flux klein", "BFL flux 2", or any explicit ask to generate with this model.

    Also in 4 projects that use this

  • agentspace-soSkill

    flux-kontext

    Edit images with Flux 1 Kontext Pro (Black Forest Labs' precise local image-edit model) on RunComfy — bundled with the model's documented prompting patterns so the skill gets sharper output than naive prompting against the same model. Documents Flux Kontext's strengths (single-reference precise local edits, strong prompt control, consistent high-fidelity outputs), the schema (single image + prompt), and when to route to Nano Banana Edit / GPT Image 2 edit / Flux 2 Klein instead. Calls `runcomfy run blackforestlabs/flux-1-kontext/pro/edit` through the local RunComfy CLI. Triggers on "flux kontext", "flux-kontext", "flux 1 kontext", "kontext", "BFL kontext", or any explicit ask to edit with this model.

    Also in 4 projects that use this

  • prime-skillsSkill

    ace-step

    Generate, inpaint, and outpaint music with ACE Step on RunComfy via the `runcomfy` CLI. ACE Step is StepFun-AI's open-weights music foundation model — tag-driven composition (genre, mood, instruments), multilingual lyrics with section markers, 5 s to 4 min stereo output, $0.0002–0.0003 per second (≈ 27× cheaper than ElevenLabs Music). Four endpoints: ACE Step text-to-audio (the default), ACE Step 1.5 text-to-audio (50+ language lyrics, refined structured-lyric handling), ACE Step audio-inpaint (regenerate a time range inside an existing track), ACE Step audio-outpaint (extend an existing track before or after). Triggers on "ace step", "ace-step", "acestep", "ACE music", "open music model", "cheap AI music", "inpaint audio", "audio inpaint", "extend music", "audio outpaint", "lengthen track", "music with tags", or any explicit ask to generate or edit music with ACE Step.

    Also in 3 projects that use this

  • prime-skillsSkill

    codex-pet

    Codex Pet generator on RunComfy. Build a Codex-compatible Codex Pet spritesheet.webp + pet.json from a single reference image, drop it into `${CODEX_HOME:-$HOME/.codex}/pets/<name>/` and Codex picks it up as a custom Codex Pet next to the 8 built-ins. This skill produces the exact Codex Pet atlas Codex expects (1536x1872 PNG/WebP, 8 cols x 9 rows, 192x208 cells, 9 animation states — idle, running-right, running-left, waving, jumping, failed, waiting, running, review). Calls OpenAI GPT Image 2 edit ONCE via the local RunComfy CLI as `runcomfy run openai/gpt-image-2/edit` to produce a canonical Codex Pet pose, then assembles all 9 animation rows programmatically with ImageMagick micro-transforms — no Codex Pro, no `$imagegen`, no OPENAI_API_KEY required, only RUNCOMFY_TOKEN. Triggers on "codex pet", "create codex pet", "make codex pet", "hatch codex pet", "/hatch image", "desktop pet codex", "codex pets", "spritesheet.webp", or any explicit ask to build a custom pet for OpenAI Codex.

    Also in 3 projects that use this

  • +3 more from agentspace-so
  • +5 more from prime-skills

The file

We split this skill into the rules it gives an agent

9 rules. 0 are common, 6 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 9 edited versions we read keep these.

  • 6

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

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

    line 12
  2. Distinctive:

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

    line 24
  3. Very specific:

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

    line 31
  4. Very specific:

    no resolution mention → 1K

    line 43
  5. Very specific:

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

    line 47
  6. Distinctive:

    If unspecified, let the model default behavior apply.

    line 62
  7. Distinctive:

    Use one image for simple edits.

    line 71
  8. Distinctive:

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

    line 79
  9. Distinctive:

    Do not read the output image back unless explicitly requested

    line 98

“Common” is measured: kept by at least 6 in 10 of the 9 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
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

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.