by 安全扫描
OpenClaw
可疑
medium confidenceThe skill's instructions match its stated purpose (calling AceDataCloud's NanoBanana image API) but the package metadata omits an environment variable the SKILL.md requires and the optional pip package is unreviewed — these inconsistencies deserve caution.
评估建议
This skill appears to do what it says (call AceDataCloud's NanoBanana image API). Before installing or using it: 1) Note the SKILL.md requires ACEDATACLOUD_API_TOKEN, but the registry metadata omitted that — treat that as a metadata error and avoid pasting tokens into unknown UIs until you confirm where the token will be stored and used. 2) Be aware that using the skill will send prompts and image URLs to https://api.acedata.cloud — do not send sensitive images or PII unless you trust the servic...详细分析 ▾
ℹ 用途与能力
The name/description (generate/edit images via AceDataCloud NanoBanana) aligns with the curl examples and editing workflows in SKILL.md. Requesting an ACEDATACLOUD_API_TOKEN is coherent for this purpose. However, the registry metadata claims 'Required env vars: none' and 'Primary credential: none', which contradicts the SKILL.md that explicitly requires ACEDATACLOUD_API_TOKEN.
✓ 指令范围
The runtime instructions are narrowly scoped to calling AceDataCloud endpoints (https://api.acedata.cloud/…) and optional installation/use of an MCP helper package. The SKILL.md does not instruct reading unrelated files, other environment variables, or sending data to unexpected third parties.
ℹ 安装机制
There is no install spec (instruction-only), which reduces risk. The documentation suggests 'pip install mcp-nano-banana' as an optional helper; installing third-party Python packages can run arbitrary code, and the package origin isn't verified in the SKILL.md. This is expected for optional tooling but worth checking before installing.
⚠ 凭证需求
The SKILL.md requires ACEDATACLOUD_API_TOKEN (appropriate for calling the API), but the skill metadata lists no required environment variables or primary credential. The SKILL.md referencing an env var that isn't declared in the registry is a manifest inconsistency and could lead to unexpected behavior or user confusion. Ensure the token scope and privileges are minimized before providing it.
✓ 持久化与权限
The skill is instruction-only, has always:false, and does not request persistent system privileges or attempt to modify other skills or global agent settings. Autonomous invocation is allowed (platform default) but does not combine with other high-risk flags here.
安全有层次，运行前请审查代码。
运行时依赖
无特殊依赖
版本
latestv1.0.02026/3/22
Initial release of nano-banana-image skill. - Generate and edit AI images via AceDataCloud NanoBanana (Gemini-based) API. - Supports both standard (`nano-banana`) and high-quality (`nano-banana-pro`) models. - Enables text-to-image and image editing workflows using natural language prompts. - No mask needed for editing; simply describe desired changes. - Returns direct image URLs with flexible aspect ratios. - Requires ACEDATACLOUD_API_TOKEN for authentication.
● 无害
安装命令 点击复制
官方npx clawhub@latest install acedatacloud-nano-banana-image
镜像加速npx clawhub@latest install acedatacloud-nano-banana-image --registry https://cn.clawhub-mirror.com
技能文档
Generate and edit AI images through AceDataCloud's NanoBanana (Gemini-based) API.
Authentication
export ACEDATACLOUD_API_TOKEN="your-token-here"
Quick 开始
curl -X POST https://api.acedata.cloud/nano-banana/images \
-H "Authorization: Bearer $ACEDATACLOUD_API_TOKEN" \
-H "Content-Type: application/json" \
-d '{"prompt": "a watercolor painting of a French countryside village", "model": "nano-banana"}'
Models
| Model | Best For |
|---|---|
nano-banana | Standard image generation (default) |
nano-banana-pro | Higher quality, more detailed output |
Workflows
1. Text-到-Image
POST /nano-banana/images
{
"prompt": "a photorealistic macro shot of morning dew on a spider web",
"model": "nano-banana-pro",
"aspect_ratio": "16:9"
}
2. Image Editing
Edit an existing image using natural language instructions — no mask needed.
POST /nano-banana/images/edit
{
"prompt": "change the background to a starry night sky",
"image_url": "https://example.com/photo.jpg",
"model": "nano-banana"
}
Parameters
Generation
| Parameter | Values | Description |
|---|---|---|
model | "nano-banana", "nano-banana-pro" | Model to use |
aspect_ratio | "1:1", "3:4", "4:3", "9:16", "16:9" | Output aspect ratio |
Editing
| Parameter | Required | Description |
|---|---|---|
image_url | Yes | URL of the source image |
prompt | Yes | Natural language editing instruction |
model | No | Model to use (defaults to "nano-banana") |
aspect_ratio | No | Output aspect ratio |
MCP Server
pip install mcp-nano-banana
Or hosted: https://nano-banana.mcp.acedata.cloud/mcp
Key tools: nano_banana_generate_image, nano_banana_edit_image
Gotchas
- Editing 做 不 require mask — 只是 describe 更改 在...中 natural language
nano-banana-proproduces significantly 更多 detailed results 但是 costs 更多- Results return direct image URL — 否 task polling needed
- Aspect ratio uses colon notation (e.g.,
"16:9") 不 pixel dimensions - Gemini-based 模型 excels 在 understanding complex, conversational editing instructions
数据来源：ClawHub ↗ · 中文优化：龙虾技能库
OpenClaw 技能定制 / 插件定制 / 私有工作流定制
免费技能或插件可能存在安全风险，如需更匹配、更安全的方案，建议联系付费定制