/install picnow
Picnow Image Generation Skill
Picnow wraps the api.letmego.top GPT-Image-2 API. This skill handles
text-to-image and image-to-image (edit) generation.
1. Environment check
Before generating, verify the token is set:
if [ -z "$LETMEGO_API_KEY" ]; then
echo "❌ LETMEGO_API_KEY is not set."
echo "Visit https://api.letmego.top to get your 令牌, then:"
echo " export LETMEGO_API_KEY=your_token"
exit 1
fi
2. Gather parameters
Ask the user if not already provided:
| Parameter | Options | Default |
|---|---|---|
prompt |
any text description | (required) |
quality |
1k · 2k · 4k |
1k |
aspect |
square · landscape · portrait |
square |
n |
1 – 10 | 1 |
ref |
local file path (optional) | (none → text-to-image) |
Quality guide:
1k— up to 1536×1536, fast and cheap, great for previews and social media2k— up to 2048×2048, suitable for e-commerce and print4k— up to 3840×2160, landscape or portrait only (no 4k square)
3. Locate and run the generation script
# Resolve skill directory (user-global → project-local)
SKILL_DIR="$HOME/.claude/skills/picnow"
[ -d "$SKILL_DIR" ] || SKILL_DIR=".claude/skills/picnow"
node "$SKILL_DIR/scripts/generate.js" \
--prompt "PROMPT_HERE" \
--quality QUALITY \
--aspect ASPECT \
--n N
# add: --ref /path/to/image.jpg for image-to-image
The script outputs a single JSON line:
{ "urls": ["https://...cdn.../image.png"] }
4. Present results
- Display each URL as a clickable link or inline image.
- For multiple images, list them with index labels (1/3, 2/3, …).
- Offer to download, save to a project, or run another variation.
Error handling
| Exit condition | Meaning | Action |
|---|---|---|
LETMEGO_API_KEY missing |
Token not set | Guide user to picnow settings |
| HTTP 401 | Token invalid | Ask user to check or refresh token |
| HTTP 402 / quota | Insufficient balance | Direct to letmego.top recharge |
| HTTP 4xx other | Bad request | Show upstream error message |
| Network error | Connection issue | Suggest retry |
Examples
Text-to-image:
User: 帮我画一张珠宝产品主图,白色背景,高清
→ quality=2k, aspect=square
→ node "$SKILL_DIR/scripts/generate.js" \
--prompt "jewelry product hero shot, white background, studio lighting, photorealistic" \
--quality 2k --aspect square --n 1
Image-to-image:
User: 把这张图改成动漫风格 [uploads reference.jpg]
→ quality=1k, ref=reference.jpg
→ node "$SKILL_DIR/scripts/generate.js" \
--prompt "anime style illustration, vibrant colors, clean linework" \
--quality 1k --aspect square --n 1 --ref reference.jpg
Landscape banner:
User: Generate a landscape banner for a coffee brand
→ quality=1k, aspect=landscape, n=2
→ node "$SKILL_DIR/scripts/generate.js" \
--prompt "coffee brand lifestyle banner, warm tones, morning light, artisan aesthetic" \
--quality 1k --aspect landscape --n 2
- 确保已安装 OpenClaw(本地或 Docker 部署)
- 在对话框中输入安装命令:
/install picnow - 安装完成后,直接呼叫该 Skill 的名称或使用
/picnow触发 - 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
Picnow — Image Generation 是什么?
Generate high-quality images via Picnow (api.letmego.top). Activate when the user wants to create, generate, draw, or design images — including product photo... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 39 次。
如何安装 Picnow — Image Generation?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install picnow」即可一键安装,无需额外配置。
Picnow — Image Generation 是免费的吗?
是的,Picnow — Image Generation 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
Picnow — Image Generation 支持哪些平台?
Picnow — Image Generation 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。
谁开发了 Picnow — Image Generation?
由 jeek(@jeekchen)开发并维护,当前版本 v1.0.1。