Nano Banana — Nano工具

Name: Nano Banana — Nano工具
Author: Keyserkazi1

Keyserkazi1

Nano Banana — Nano工具

v1.0.0

Generate/edit images with Nano Banana 图像工具 Pro (Gemini 3 Pro Image). Use for image create/modify requests incl. edits. Supports 文生图 + 图生图; 1K...

0· 197·0 当前·0 累计

by @keyserkazi1 (Keyserkazi1)·MIT-0

开发工具 API工具代码生成网络工具

下载技能包

License

MIT-0

最后更新

2026/4/12

安全扫描

VirusTotal

无害

查看报告

OpenClaw

可疑

medium confidence

The skill's code and instructions generally match its image-generation purpose, but there are inconsistencies (undeclared required API key, missing install steps for dependencies, and metadata mismatches) that merit caution before installing.

评估建议

This skill appears to implement Gemini-based image generation and editing and does not contain obvious exfiltration. However: - The script requires a GEMINI_API_KEY (or --api-key) but the registry metadata does not declare that requirement — verify you are comfortable supplying an API key and that the key has appropriate (limited) permissions and quota. - There is no install spec: you must ensure Python >=3.10 and install google-genai and pillow before running. Consider running it in an isolated...

详细分析 ▾

⚠ 用途与能力

The skill claims to generate/edit images via Google's Gemini (Nano Banana / Gemini 3 Pro Image) and the included script implements that (uses google.genai, calls model 'gemini-3-pro-image-preview'). However the registry metadata declares no required environment variables or primary credential while the script and SKILL.md clearly require a GEMINI_API_KEY or --api-key. Also the _meta.json ownerId does not match the registry ownerId, creating an ownership/metadata inconsistency.

✓ 指令范围

SKILL.md and the script limit actions to generating or editing images and saving them to disk. The runtime instructions only reference the GEMINI API key and local image files; they do not instruct reading unrelated system files or exfiltrating data to unexpected endpoints. The script prints model text output and the saved image path, which is expected for this purpose.

ℹ 安装机制

There is no install spec despite the script listing dependencies (google-genai, pillow) in comments. That means users/agents must ensure Python >=3.10 and install those packages themselves. Absence of an install step is not malicious but is an inconsistency and could lead to runtime failures.

⚠ 凭证需求

The script requires a GEMINI_API_KEY (or --api-key) to operate, but the registry lists no required env vars or primary credential — this is a mismatch. Requesting a single API key for the external image model is proportionate to the stated purpose, but the missing declaration and lack of guidance about key scope/permissions raises concern.

✓ 持久化与权限

The skill is not always-included, is user-invocable, and does not request elevated or persistent privileges. It does not modify other skills or global agent settings.

安全有层次，运行前请审查代码。

License

MIT-0

可自由使用、修改和再分发，无需署名。

查看条款 ↗

运行时依赖

无特殊依赖

版本

latestv1.0.02026/3/17

- Initial release of Nano Banana Pro skill. - Provides image generation and editing using Google's Gemini 3 Pro Image API. - Supports text-to-image and image-to-image in 1K, 2K, and 4K resolutions. - Includes workflow guidance, filename conventions, prompt handling, and error troubleshooting. - Usage instructions and practical examples are documented.

● 无害

安装命令点击复制

官方npx clawhub@latest install nano-banana-pro1

镜像加速npx clawhub@latest install nano-banana-pro1 --registry https://cn.clawhub-mirror.com

技能文档

Generate new images or edit existing ones using Google's Nano Banana Pro API (Gemini 3 Pro Image).

Usage

Run the script using absolute path (do NOT cd to skill directory first):

Generate 新的 image:

uv run ~/.codex/skills/nano-banana-pro/scripts/generate_image.py --prompt "your image description" --filename "output-name.png" [--resolution 1K|2K|4K] [--api-key KEY]

编辑 existing image:

uv run ~/.codex/skills/nano-banana-pro/scripts/generate_image.py --prompt "editing instructions" --filename "output-name.png" --input-image "path/to/input.png" [--resolution 1K|2K|4K] [--api-key KEY]

Important: Always run 从用户's current working directory 所以 images saved 在哪里用户 working, 不在...中 skill directory.

默认 Workflow (draft → iterate → final)

Goal: fast iteration without burning time on 4K until the prompt is correct.

Draft (1K): quick feedback 循环

-

uv run ~/.codex/skills/nano-banana-pro/scripts/generate_image.py --prompt "" --filename "yyyy-mm-dd-hh-mm-ss-draft.png" --resolution 1K

Iterate: adjust prompt 在...中 small diffs; keep filename 新的 per run

- 如果 editing: keep 相同 --输入框-image 对于 every iteration until 您’re happy.

Final (4K): 仅当...时 prompt locked

-

uv run ~/.codex/skills/nano-banana-pro/scripts/generate_image.py --prompt "" --filename "yyyy-mm-dd-hh-mm-ss-final.png" --resolution 4K

Resolution Options

The Gemini 3 Pro Image API supports three resolutions (uppercase K required):

1K (默认) - ~1024px resolution
2K - ~2048px resolution
4K - ~4096px resolution

Map user requests to API parameters:

否 mention 的 resolution → 1K
"low resolution", "1080", "1080p", "1K" → 1K
"2K", "2048", "normal", "medium resolution" → 2K
"high resolution", "high-res", "hi-res", "4K", "ultra" → 4K

API 键

The script checks for API key in this order:

--api-键 argument (使用如果用户 provided 键在...中 chat)
GEMINI_API_KEY environment 变量

If neither is available, the script exits with an error message.

Preflight + Common Failures (fast fixes)

Preflight:

- command -v uv (必须 exist) - test -n \"$GEMINI_API_KEY\" (或 pass --api-键) - 如果 editing: test -f \"path/到/输入框.png\"

Common failures:

- 错误: 否 API 键 provided. → 设置 GEMINI_API_KEY 或 pass --api-键 - 错误 loading 输入框 image: → wrong path / unreadable file; 验证 --输入框-image points 到 real image - “quota/权限/403” 样式 API errors → wrong 键, 否 access, 或 quota exceeded; try 不同键/账户

Filename Generation

Generate filenames with the pattern: yyyy-mm-dd-hh-mm-ss-name.png

格式: {时间戳}-{descriptive-name}.png

时间戳: Current 日期/时间在...中格式 yyyy-mm-dd-hh-mm-ss (24-hour 格式)
Name: Descriptive lowercase text 带有 hyphens
Keep descriptive part concise (1-5 words typically)
使用 context 从用户's prompt 或 conversation
如果 unclear, 使用 random identifier (e.g., x9k2, a7b3)

Examples:

Prompt " serene Japanese garden" → 2025-11-23-14-23-05-japanese-garden.png
Prompt "sunset 在...上 mountains" → 2025-11-23-15-30-12-sunset-mountains.png
Prompt "创建 image 的 robot" → 2025-11-23-16-45-33-robot.png
Unclear context → 2025-11-23-17-12-48-x9k2.png

Image Editing

When the user wants to modify an existing image:

Check 如果它们 provide image path 或 reference image 在...中 current directory
使用 --输入框-image parameter 带有 path 到 image
prompt 应该 contain editing instructions (e.g., "使 sky 更多 dramatic", "移除 person", "更改到 cartoon 样式")
Common editing tasks: 添加/移除 elements, 更改样式, adjust colors, blur background, etc.

Prompt Handling

对于 generation: Pass 用户's image description 作为- 到 --prompt. 仅 rework 如果 clearly insufficient.

对于 editing: Pass editing instructions 在...中 --prompt (e.g., "添加 rainbow 在...中 sky", "使 look 点赞 watercolor painting")

Preserve user's creative intent in both cases.

Prompt Templates (high hit-rate)

Use templates when the user is vague or when edits must be precise.

Generation 模板:

- “创建 image 的: . 样式: <样式>. Composition: . Lighting: . Background: . Color palette: . Avoid: <列表>.”

Editing 模板 (preserve everything 否则):

- “更改仅: . Keep identical: subject, composition/crop, pose, lighting, color palette, background, text, 和 overall 样式. 做不添加新的 objects. 如果 text exists, keep unchanged.”

输出

Saves PNG 到 current directory (或 specified path 如果 filename includes directory)
Script outputs 满 path 到 generated image
做不读取 image back - 只是 inform 用户的 saved path

Examples

Generate 新的 image:

uv run ~/.codex/skills/nano-banana-pro/scripts/generate_image.py --prompt "A serene Japanese garden with cherry blossoms" --filename "2025-11-23-14-23-05-japanese-garden.png" --resolution 4K

编辑 existing image:

uv run ~/.codex/skills/nano-banana-pro/scripts/generate_image.py --prompt "make the sky more dramatic with storm clouds" --filename "2025-11-23-14-25-30-dramatic-sky.png" --input-image "original-photo.jpg" --resolution 2K

Generate new images or edit existing ones using Google's Nano Banana Pro API (Gemini 3 Pro Image).

Usage

Run the script using absolute path (do NOT cd to skill directory first):

Generate new image:

uv run ~/.codex/skills/nano-banana-pro/scripts/generate_image.py --prompt "your image description" --filename "output-name.png" [--resolution 1K|2K|4K] [--api-key KEY]

Edit existing image:

uv run ~/.codex/skills/nano-banana-pro/scripts/generate_image.py --prompt "editing instructions" --filename "output-name.png" --input-image "path/to/input.png" [--resolution 1K|2K|4K] [--api-key KEY]

Important: Always run from the user's current working directory so images are saved where the user is working, not in the skill directory.

Default Workflow (draft → iterate → final)

Goal: fast iteration without burning time on 4K until the prompt is correct.

Draft (1K): quick feedback loop

-

uv run ~/.codex/skills/nano-banana-pro/scripts/generate_image.py --prompt "" --filename "yyyy-mm-dd-hh-mm-ss-draft.png" --resolution 1K

Iterate: adjust prompt in small diffs; keep filename new per run

- If editing: keep the same --input-image for every iteration until you’re happy.

Final (4K): only when prompt is locked

-

uv run ~/.codex/skills/nano-banana-pro/scripts/generate_image.py --prompt "" --filename "yyyy-mm-dd-hh-mm-ss-final.png" --resolution 4K

Resolution Options

The Gemini 3 Pro Image API supports three resolutions (uppercase K required):

1K (default) - ~1024px resolution
2K - ~2048px resolution
4K - ~4096px resolution

Map user requests to API parameters:

No mention of resolution → 1K
"low resolution", "1080", "1080p", "1K" → 1K
"2K", "2048", "normal", "medium resolution" → 2K
"high resolution", "high-res", "hi-res", "4K", "ultra" → 4K

API Key

The script checks for API key in this order:

--api-key argument (use if user provided key in chat)
GEMINI_API_KEY environment variable

If neither is available, the script exits with an error message.

Preflight + Common Failures (fast fixes)

Preflight:

- command -v uv (must exist) - test -n \"$GEMINI_API_KEY\" (or pass --api-key) - If editing: test -f \"path/to/input.png\"

Common failures:

- Error: No API key provided. → set GEMINI_API_KEY or pass --api-key - Error loading input image: → wrong path / unreadable file; verify --input-image points to a real image - “quota/permission/403” style API errors → wrong key, no access, or quota exceeded; try a different key/account

Filename Generation

Generate filenames with the pattern: yyyy-mm-dd-hh-mm-ss-name.png

Format: {timestamp}-{descriptive-name}.png

Timestamp: Current date/time in format yyyy-mm-dd-hh-mm-ss (24-hour format)
Name: Descriptive lowercase text with hyphens
Keep the descriptive part concise (1-5 words typically)
Use context from user's prompt or conversation
If unclear, use random identifier (e.g., x9k2, a7b3)

Examples:

Prompt "A serene Japanese garden" → 2025-11-23-14-23-05-japanese-garden.png
Prompt "sunset over mountains" → 2025-11-23-15-30-12-sunset-mountains.png
Prompt "create an image of a robot" → 2025-11-23-16-45-33-robot.png
Unclear context → 2025-11-23-17-12-48-x9k2.png

Image Editing

When the user wants to modify an existing image:

Check if they provide an image path or reference an image in the current directory
Use --input-image parameter with the path to the image
The prompt should contain editing instructions (e.g., "make the sky more dramatic", "remove the person", "change to cartoon style")
Common editing tasks: add/remove elements, change style, adjust colors, blur background, etc.

Prompt Handling

For generation: Pass user's image description as-is to --prompt. Only rework if clearly insufficient.

For editing: Pass editing instructions in --prompt (e.g., "add a rainbow in the sky", "make it look like a watercolor painting")

Preserve user's creative intent in both cases.

Prompt Templates (high hit-rate)

Use templates when the user is vague or when edits must be precise.

Generation template:

- “Create an image of: . Style:

License

运行时依赖

版本

安装命令 点击复制

技能文档

Usage

默认 Workflow (draft → iterate → final)

Resolution Options

API 键

Preflight + Common Failures (fast fixes)

Filename Generation

Image Editing

Prompt Handling

Prompt Templates (high hit-rate)

输出

Examples

Usage

Default Workflow (draft → iterate → final)

Resolution Options

API Key

Preflight + Common Failures (fast fixes)

Filename Generation

Image Editing

Prompt Handling

Prompt Templates (high hit-rate)

安装命令点击复制