← 返回 Skills 市场
550
总下载
1
收藏
1
当前安装
10
版本数
在 OpenClaw 中安装
/install nano-banana-pro-image
功能描述
Generate and edit images using Google's Nano Banana Pro (Gemini 3 Pro Image / Imagen Pro) — the premium AI image generation model optimized for professional...
安全使用建议
This skill appears coherent and limited to Atlas Cloud image operations, but before installing: 1) Only provide ATLASCLOUD_API_KEY and keep it minimal/rotatable; avoid putting long-lived keys in shared environments. 2) Be cautious uploading sensitive local images — the script will upload files to Atlas Cloud after an explicit confirmation. 3) Review the included script if you require additional guarantees (e.g., TLS verification policy, logging behavior). 4) Run it in a least-privileged environment and monitor network activity if you want extra assurance that traffic only targets api.atlascloud.ai. If any part of the code was modified from the repository, re-audit changes before use.
功能分析
Type: OpenClaw Skill
Name: nano-banana-pro-image
Version: 1.0.9
The skill bundle provides a legitimate interface for professional image generation and editing via the Atlas Cloud API. The core logic in `scripts/generate_image.py` is well-structured, uses only Python standard libraries, and communicates exclusively with the official `api.atlascloud.ai` endpoint. It includes security-conscious features such as reading credentials from environment variables and requiring user confirmation before uploading local files. The `SKILL.md` instructions are aligned with the script's functionality and do not contain any prompt-injection attempts or malicious directives.
能力评估
Purpose & Capability
Name/description, SKILL.md, and the included script all describe image generation/editing via Atlas Cloud and require only ATLASCLOUD_API_KEY. Required env var and declared primary credential match the service used (api.atlascloud.ai). No unrelated credentials, binaries, or config paths are requested.
Instruction Scope
SKILL.md and the Python script limit operations to: submitting generation/edit requests, polling predictions, downloading result files, listing models, and optionally uploading user-confirmed local files. The SKILL.md explicitly instructs using JSON request bodies and headers (not shell interpolation) and the script reads only ATLASCLOUD_API_KEY and user-supplied file paths when uploading. The skill does save downloaded images to disk and will read local files only when the user explicitly invokes upload.
Install Mechanism
No install spec is provided (instruction-only skill with one helper script). Nothing is downloaded from untrusted URLs and there are no extract-or-execute install steps. The included Python script uses only the standard library.
Credentials
Only one environment variable (ATLASCLOUD_API_KEY) is required and it directly maps to the Atlas Cloud API used by the skill. No extraneous secrets or unrelated service credentials are requested.
Persistence & Privilege
always is false and the skill does not request permanent platform-level privileges or modify other skills. The agent may invoke the skill autonomously (platform default), which is expected for a service integration.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install nano-banana-pro-image - 安装完成后，直接呼叫该 Skill 的名称或使用
/nano-banana-pro-image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
版本历史
v1.0.9
nano-banana-pro-image 1.0.9
- No code or documentation changes detected in this version.
- Functionality and documentation remain unchanged from the previous release.
v1.0.8
No functional changes; documentation now reflects Atlas Cloud as the only supported provider for this skill.
- Removed references to Google AI Studio/Gemini API support and setup.
- Clarified that all image generation and editing now use the Atlas Cloud API.
- Updated examples, parameters, and workflow to only reference Atlas Cloud endpoints and key usage.
- Revised pricing/comparison tables to remove Google-specific data.
v1.0.7
- No code or documentation changes detected in this release.
- Version updated with no modifications to functionality or files.
v1.0.6
- Dropped requirement for the GEMINI_API_KEY environment variable; now only ATLASCLOUD_API_KEY is required.
- Updated metadata to reflect the change in required environment variables.
- No changes to functionality or features.
v1.0.5
nano-banana-pro-image 1.0.5
- Added OpenClaw metadata to SKILL.md, specifying required environment variables for API access.
- No changes to code or functionality.
v1.0.4
- Changed environment variable requirements: ATLASCLOUD_API_KEY is now required by default, with GEMINI_API_KEY as an alternative (not required).
- Updated descriptions in SKILL.md to clarify provider setup and API key requirements.
- Clarified that Atlas Cloud API key is needed unless using Google AI Studio.
- No functional changes to the core generation or editing features.
v1.0.3
- Introduced a new Python script (`scripts/generate_image.py`) for seamless image generation, model listing, and image upload/editing from the command line.
- Updated documentation to include script usage examples for generating, uploading, and editing images.
- No breaking changes; existing API and usage remain unchanged.
- Adds convenience for users preferring command line tools over direct API calls.
v1.0.2
- Added a security note clarifying API key handling and safe practices for constructing API requests.
- Documented that API keys are passed via HTTP headers and only read from environment variables, not embedded in URLs or command arguments.
- Updated instructions to require using JSON request bodies for all user-supplied values to prevent shell injection.
- Clarified that the skill does not execute user code and only performs structured API requests and downloads.
- No functional changes to the skill — documentation update only.
v1.0.1
**Nano Banana Pro replaces Nano Banana 2 with professional-grade image generation.**
- Upgraded to "Nano Banana Pro" (Gemini 3 Pro Image/Imagen Pro) for advanced reasoning, superior text rendering, and complex multi-object composition.
- Supports text-to-image and image editing with up to 6 reference images, more aspect ratios, and up to 4K resolution.
- New models and APIs: `gemini-3-pro-image-preview` (Google) and `google/nano-banana-pro/*` (Atlas Cloud), with updated pricing details.
- Clarified provider selection and setup instructions for both Atlas Cloud and Google AI Studio.
- Documentation now emphasizes professional use cases: marketing assets, infographics, menus, and other text-focused high-quality visuals.
- Data privacy notice: Alerts users that their data is sent to third-party APIs for processing.
v1.0.0
Major update: Skill is now upgraded to Nano Banana 2 (Imagen) model with multiple provider options, improved pricing, and expanded capabilities.
- Migrated from Nano Banana Pro (Gemini 3 Pro Image) to Nano Banana 2 (Imagen), supporting higher quality image generation and editing.
- Added support for both Atlas Cloud (recommended, cheaper, multi-provider platform) and Google AI Studio (official Gemini API) — user can choose.
- Updated documentation for new API endpoints, workflow (job submission, polling), and multiple reference image editing (up to 14 images).
- Included comprehensive pricing comparison, feature overview, and step-by-step usage/setup for both providers.
- Enhanced skill triggers — now recommended for any AI image/art/visual asset generation (sprites, banners, logos, etc.), not just Nano Banana.
- Removed legacy script-based workflow and detailed Bash instructions for the old model.
元数据
常见问题
Nano Banana Pro Image Generation&Editing 是什么？
Generate and edit images using Google's Nano Banana Pro (Gemini 3 Pro Image / Imagen Pro) — the premium AI image generation model optimized for professional... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 550 次。
如何安装 Nano Banana Pro Image Generation&Editing？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install nano-banana-pro-image」即可一键安装，无需额外配置。
Nano Banana Pro Image Generation&Editing 是免费的吗？
是的，Nano Banana Pro Image Generation&Editing 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Nano Banana Pro Image Generation&Editing 支持哪些平台？
Nano Banana Pro Image Generation&Editing 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 Nano Banana Pro Image Generation&Editing？
由 MikeWang（@xixihhhh）开发并维护，当前版本 v1.0.9。
推荐 Skills