Agent skill

Nano Banana

by nexu-io in nexu-io/nexu

Generate or edit images via Nano Banana image models. An agent skill from nexu-io/nexu.

MITAuto-check passedMedia & Creative

Install Nano Banana

skills CLI
$ npx skills add nexu-io/nexu --skill nano-banana -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install nexu-io/nexu nano-banana --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/nexu-io/nexu.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/nexubot/nano-banana .claude/skills/nano-banana && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
nano-banana
GitHub stars
3.3k
Token cost
~757 tokens
SKILL.md length
268 words
Files
3 (incl. scripts)
Skills in repo
10
Repo updated
First seen
Licence
MIT

At a glance

Generate or edit images via Nano Banana image models. An agent skill from nexu-io/nexu.

  • Image generation
  • SKILL.md covers Models, Generate an image, Edit a single image and Multi-image composition (up to…, plus 5 more sections
  • Runs JavaScript scripts from its folder; calls node
  • Nano banana pro

What it does

Nano Banana is an agent skill from nexu-io/nexu. Generate or edit images via Nano Banana image models. Triggers on "generate image", "image generation", "nano banana", "edit image", "nano banana pro", "nano banana 2"

Its SKILL.md is about 760 tokens, which your agent loads only when the skill is triggered. The skill folder holds 3 other files, including scripts (for example `scripts/generate-image.js` and `scripts/package.json`).

It sits in Media & Creative, covering Image generation and Image editing. It works with Google Gemini, WeChat, Slack and Discord. The repository describes itself as: The simplest desktop client for OpenClaw 🦞 — bridge your Agent to WeChat, Feishu, Slack & Discord in one click. Works with Claude Code, Codex & any LLM. BYOK, Oauth… The licence is MIT.

When your agent uses it

  • Image generation
  • Nano banana pro

Example prompts

  • “generate image”
  • “image generation”
  • “nano banana”
  • “/nano-banana”

Requirements

  • Node.js

What it can do on your machine

Read from SKILL.md and the folder at commit dadfb1c. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 2 files in scripts/ (JavaScript), which the agent can run.

    Shell commands in SKILL.md call:

    • node

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Nano Banana loads about 757 tokens when it runs. Until then it costs about 45 tokens; SKILL.md has 268 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~45
When it runs · the whole SKILL.md, loaded when a task matches
~757

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from nexu-io/nexu at commit dadfb1c, republished under its MIT licence (© nexu-io). 268 words, ~757 tokens.

Download SKILL.mdSave it as .claude/skills/nano-banana/SKILL.md (or your agent's skills folder). This skill also uses 2 other files; get the full folder from GitHub.
name
nano-banana
description
Generate or edit images via Nano Banana image models. Triggers on "generate image", "image generation", "nano banana", "edit image", "nano banana pro", "nano banana 2"
homepage
https://ai.google.dev/

Nano Banana — Image Generation

Image generation script supporting three models. Requires sharp for input image compression (auto-installed on first run).

Models

FlagNotes
--model nano-bananaDefault. Fast, good quality.
--model nano-banana-proHighest quality, slower.
--model nano-banana-2Legacy model.

Generate an image

bash
node {baseDir}/scripts/generate-image.js --prompt "a cat sitting on mars" --filename "cat-on-mars.png"

Edit a single image

bash
node {baseDir}/scripts/generate-image.js \
  --prompt "make the sky purple" \
  --filename "edited.png" \
  -i "/path/to/input.png" \
  --model nano-banana-pro

Multi-image composition (up to 14 images)

bash
node {baseDir}/scripts/generate-image.js \
  --prompt "combine these into a collage" \
  --filename "collage.png" \
  -i img1.png -i img2.png -i img3.png

Options

FlagShortDefaultDescription
--prompt-prequiredImage description or editing instruction
--filename-frequiredOutput filename
--input-image-i—Input image(s), repeatable, max 14
--model—nano-banananano-banana, nano-banana-pro, or nano-banana-2
--resolution-r1K1K, 2K, or 4K
--aspect-ratio——e.g. 1:1, 16:9, 4:3, 3:4, 9:16

API key

The API key is pre-configured on this machine. No flags or environment variables needed.

Input image handling

All input images are sent as inline base64. Images over 500 KB are automatically compressed to JPEG and resized to fit under the limit. This keeps requests fast and avoids File API auth issues with the enterprise endpoint.

Output

Relative filenames are saved to $OPENCLAW_STATE_DIR/media/outbound/{slugid}/nano-banana/{filename}. Absolute paths are used as-is. Absolute paths are used as-is. Use timestamps in filenames to avoid overwrites: cat-on-mars-20260304-165000.png.

Sending images to the user

The script prints a MEDIA: <absolute-path> line on stdout. You MUST include this exact MEDIA: line in your reply text so the image is delivered as an attachment in Discord/Slack/chat.

Example reply:

Here's your image!
MEDIA: /Users/alche/.openclaw/media/outbound/my-bot/nano-banana/cat-on-mars.png

Rules:

  • Copy the MEDIA: line from the script output into your reply verbatim — this is how images get sent
  • Do NOT read the generated image back with the read tool
  • Do NOT try to base64 encode or manually attach the image
  • The MEDIA: line must be on its own line in your response

© nexu-io, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 2 other files (scripts) in skills/nexubot/nano-banana of nexu-io/nexu.

  • SKILL.md
  • scripts/generate-image.js
  • scripts/package.json

Open the folder on GitHubat commit dadfb1c

Compare with similar skills

Nano Banana next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Nano Banana compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Nano Banana this skillnexu-io/nexu3.3k—~757Automated safety check: PassMIT
Use Avibeavibe-bot/avibe624—~2.5kAutomated safety check: PassMIT
Chatgpt Image Adkrusemediallc/arcads-claude-code1.6k—~2.7kAutomated safety check: NotesMIT
Nano Banana Proswarmclawai/swarmclaw689—~481Automated safety check: PassMIT
Atlas Cloudcalesthio/OpenMontage66k—~1.2kAutomated safety check: PassAGPL-3.0
AI Image Creatorevolution-foundation/evo-nexus545—~5.1kAutomated safety check: NotesCustom licence

Similar skills

  • Use Avibe

    avibe-bot/avibe

    Safely inspect and modify local Avibe configuration, routing, runtime settings, watches, scheduled tasks, Avibe Cloud remote access, and operational state.

    624 GitHub stars~2.5k tokensUpdated today
    Productivity & AutomationAuto-check passed
  • Chatgpt Image Ad

    krusemediallc/arcads-claude-code

    Generate one or more standalone Meta image-ad creatives via ChatGPT Image 2 (gpt-image-2) through the Arcads external API.

    1.6k GitHub stars~2.7k tokensUpdated 18 days ago
    Media & CreativeAuto-check: notes
  • Nano Banana Pro

    swarmclawai/swarmclaw

    Generate or edit images via Gemini 3 Pro Image (Nano Banana Pro).

    689 GitHub stars~481 tokensUpdated 3 mo ago
    Media & CreativeAuto-check passed
  • Atlas Cloud

    calesthio/OpenMontage

    Generate or edit images and videos through the Atlas Cloud gateway.

    66k GitHub stars~1.2k tokensUpdated 7 days ago
    Media & CreativeAuto-check passed
  • AI Image Creator

    evolution-foundation/evo-nexus

    Generates PNG images through OpenRouter models, with transparent backgrounds and reference-image edits, and describes existing images with multimodal vision.

    545 GitHub stars~5.1k tokensUpdated 5 mo ago
    Media & CreativeAuto-check: notes
  • Generate Image

    K-Dense-AI/claude-scientific-writer

    Generate or edit images with AI models through the OpenRouter Image API (Gemini, Seedream, Recraft, GPT-Image, Riverflow).

    2.4k GitHub starsUsed in 1 repo~3.8k tokens
    Media & CreativeAuto-check: notes

More from nexu-io/nexu

All 10 skills in this repo
  • Libtv Video

    nexu-io/nexu

    Seedance 2.0 video & image generation via LibTV Gateway - AI text-to-video, image-to-video, video continuation, style transfer, and text-to-image using Seedance 2.0 model.

    3.3k GitHub stars~3.5k tokensUpdated 5 mo ago
    Auto-check passed
  • All-in-one image generation with Gemini models. An agent skill from nexu-io/nexu.

    3.3k GitHub stars~870 tokensUpdated 5 mo ago
    Auto-check passed
  • 一句话生成大师级海报、书籍封面、专辑封面和各类设计作品。无需懂PS、配色或艺术史,AI自动选择最佳风格(基于33+位传奇设计师)。支持多平台多比例:公众号封面(21:9)、小红书配图(3:4)、文章配图(16:9)、书籍封面(9:16)、专辑封面(1:1)、电影海报(9:16)。包含AI提示词优化、风格对比、图生图转换功能。触发词:"Mondo风格"、"书籍封面设计"、"专辑封面"、"海报设计"…

    3.3k GitHub starsUsed in 1 repo~4.1k tokens
    Auto-check passed
  • Static Deploy

    nexu-io/nexu

    Deploy static pages to nexu.space. An agent skill from nexu-io/nexu.

    3.3k GitHub stars~989 tokensUpdated 5 mo ago
    Auto-check passed
  • Nexu E2E Test

    nexu-io/nexu

    A skill your agent uses when verifying OpenClaw gateway fixes end-to-end, testing skill loading after restart, or running integration tests against the local Nexu+OpenClaw stack.

    3.3k GitHub stars~1.7k tokensUpdated 5 mo ago
    Auto-check passed
  • Process PR Reviews

    nexu-io/nexu

    A skill your agent uses when the user asks to process, triage, fetch, view, count, list, or resolve review feedback in a GitHub PR.

    3.3k GitHub stars~2.5k tokensUpdated 5 mo ago
    Auto-check passed

Questions about Nano Banana

What does Nano Banana do?

Generate or edit images via Nano Banana image models. An agent skill from nexu-io/nexu. Nano Banana is an agent skill from nexu-io/nexu. Generate or edit images via Nano Banana image models.

When should I use Nano Banana?

Nano Banana fits situations like: image generation; nano banana pro.

How do I install Nano Banana in Claude Code?

Run `npx skills add nexu-io/nexu --skill nano-banana -a claude-code`. Or copy the skill folder (skills/nexubot/nano-banana in nexu-io/nexu) into .claude/skills/nano-banana in your project. Claude Code loads it when a task matches its description.

How do I install Nano Banana in Codex?

Run `npx skills add nexu-io/nexu --skill nano-banana -a codex`. Or copy the skill folder (skills/nexubot/nano-banana in nexu-io/nexu) into .agents/skills/nano-banana in your project. Codex loads it when a task matches its description.

Can I use Nano Banana in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add nexu-io/nexu --skill nano-banana -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/nano-banana, .gemini/skills/nano-banana, .github/skills/nano-banana and .opencode/skills/nano-banana in your project.

What does Nano Banana need to run?

Going by SKILL.md and its folder, Nano Banana needs JavaScript for the scripts in its folder and the command-line tools its instructions call (node). Our summary lists: Node.js.

Does Nano Banana access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Nano Banana safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Nano Banana use?

Nano Banana is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Nano Banana use?

About 757 tokens (SKILL.md is roughly 3k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Nano Banana?

Skills that share tags, products or a category with Nano Banana: Use Avibe (avibe-bot/avibe, 624 stars), Chatgpt Image Ad (krusemediallc/arcads-claude-code, 1.6k stars), Nano Banana Pro (swarmclawai/swarmclaw, 689 stars) and Atlas Cloud (calesthio/OpenMontage, 66k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Nano Banana?

nexu-io (a GitHub organization) maintains it in nexu-io/nexu, which has 3,284 GitHub stars. The repository holds 10 skills in this directory. The repository was last updated on April 26, 2026.

Source: nexu-io/nexu on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.