Back to Registry View Author Profile
Official Verified
nano-banana-pro-custom
Generate or edit images using OpenAI-compatible API. Supports multi-image input, fine-tuned models, and multiple configuration sources (env vars, openclaw.json, config.json). Use when generating images, editing images, or composing multiple images with OpenAI-style image APIs.
skill-install — Terminal
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/chenxchen/nano-banana-pro-customOr
Nano Banana Pro (OpenAI-Compatible Image Generation)
使用 OpenAI 兼容 API 生成或编辑图片,支持多图输入、微调模型和多源配置。
配置方式(优先级从高到低)
- 命令行参数
--base-url,--api-key,--model - 环境变量
NANO_BASE_URL,NANO_API_KEY,NANO_MODEL - openclaw.json
~/.openclaw/openclaw.json中的skills.entries.nano-banana-pro-customapiKey字段用于配置 API Keyenv字段用于配置其他环境变量
- 技能 config.json
{baseDir}/config.json
环境变量配置
export NANO_BASE_URL="https://api.openai.com/v1"
export NANO_API_KEY="your-api-key"
export NANO_MODEL="gpt-image-1"
openclaw.json 配置
使用命令行设置(推荐)
# 设置 API Key(特殊字段)
openclaw config set skills.entries.nano-banana-pro-custom.apiKey "sk-your-api-key"
# 设置 Base URL(env 字段)
openclaw config set skills.entries.nano-banana-pro-custom.env.NANO_BASE_URL "https://api.openai.com/v1"
# 设置模型(env 字段)
openclaw config set skills.entries.nano-banana-pro-custom.env.NANO_MODEL "gpt-image-1"
OpenRouter 示例
# OpenRouter 配置
openclaw config set skills.entries.nano-banana-pro-custom.env.NANO_BASE_URL "https://openrouter.ai/api/v1"
openclaw config set skills.entries.nano-banana-pro-custom.env.NANO_MODEL "google/gemini-3.1-flash-image-preview"
手动编辑文件
也可以直接编辑 ~/.openclaw/openclaw.json:
{
"skills": {
"entries": {
"nano-banana-pro-custom": {
"apiKey": "sk-your-api-key",
"env": {
"NANO_BASE_URL": "https://openrouter.ai/api/v1",
"NANO_MODEL": "google/gemini-3.1-flash-image-preview"
}
}
}
}
}
技能 config.json 配置
在技能目录下创建 config.json 文件:
# 创建配置文件
cat > {baseDir}/config.json << 'EOF'
{
"baseUrl": "https://api.openai.com/v1",
"apiKey": "your-api-key",
"model": "gpt-image-1"
}
EOF
配置文件内容示例:
{
"baseUrl": "https://api.openai.com/v1",
"apiKey": "your-api-key",
"model": "gpt-image-1"
}
查看当前配置
uv run {baseDir}/scripts/generate_image.py --show-config
快速开始
生成新图片
uv run {baseDir}/scripts/generate_image.py \
--prompt "一只宇航员猫在月球上" \
--output "cat_astronaut.png"
编辑单张图片
uv run {baseDir}/scripts/generate_image.py \
--prompt "给这只猫加上一顶帽子" \
--input cat.png \
--output "cat_with_hat.png"
多图合成
uv run {baseDir}/scripts/generate_image.py \
--prompt "将这两张图片融合成一个场景" \
--input img1.png \
--input img2.png \
--output "merged.png"
使用微调模型
uv run {baseDir}/scripts/generate_image.py \
--prompt "以我训练的风格画一辆车" \
--output "car_custom.png" \
--model "ft:image-model:my-finetuned-model"
参数说明
Metadata
AI Skill Finder
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skill Add to Configuration
Paste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-chenxchen-nano-banana-pro-custom": {
"enabled": true,
"auto_update": true
}
}
}Safety NoteClawKit audits metadata but not runtime behavior. Use with caution.