← 返回 Skills 市场
305
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install clawhub-publish-146230
功能描述
Generate images using the internal Google Antigravity API (Gemini 3 Pro Image). High quality, native generation without browser automation.
使用说明 (SKILL.md)
\r \r
Antigravity Image Generation\r
\r
Generate high-quality images using the internal Google Antigravity API (Gemini 3 Pro Image). This skill bypasses the need for browser automation by using the daily-cloudcode-pa.sandbox endpoint directly with your OAuth credentials.\r
\r
Prerequisites\r
\r
- Google Antigravity OAuth Profile: Must be present in your OpenClaw auth-profiles.json.\r
- Node.js: Available in the environment.\r
- Security Note: This skill reads local OAuth tokens from your profile to authenticate with Google's API. This is expected behavior for internal tool use.\r \r
Usage\r
\r
Direct Script Execution\r
\r
/home/ubuntu/clawd/skills/antigravity-image-gen/scripts/generate.js \\r
--prompt "A futuristic city on Mars" \\r
--output "/tmp/mars.png" \\r
--aspect-ratio "16:9"\r
```\r
\r
### Arguments\r
\r
- `--prompt` (Required): The description of the image.\r
- `--output` (Optional): Path to save the image (default: `/tmp/antigravity_\x3Cts>.png`).\r
- `--aspect-ratio` (Optional): `1:1` (default), `16:9`, `9:16`, `4:3`, `3:4`.\r
\r
## Output\r
\r
- The script writes the image to the specified path.\r
- It prints `MEDIA: \x3Cpath>` to stdout, which allows Clawdbot to automatically detect and display the image.\r
\r
## Troubleshooting\r
\r
- **429 Resource Exhausted**: Quota limit reached. Wait or check your project limits.\r
- **No image data found**: The model might have refused the prompt (safety) or the API structure changed. Check the "Model message" output.\r
- **Auth Error**: Ensure you have logged in via `google-antigravity` provider.\r
安全使用建议
Proceed cautiously. The SKILL.md says it will read your local Google OAuth profile and call an internal/sandbox Google endpoint, but the skill package contains no script code (the generate.js it references is missing) and the source is unknown. Before installing or using it: 1) ask the publisher for the actual script/source and review it so you can confirm it only uses your tokens to call the expected Google API; 2) verify which auth.profiles file it will read and avoid using high‑privilege or production credentials (use a limited sandbox account if possible); 3) if you cannot review the code, run it in an isolated environment or decline installation; and 4) prefer skills with verifiable source/homepage and included code or official provider plugins.
功能分析
Type: OpenClaw Skill
Name: clawhub-publish-146230
Version: 2.0.0
The skill bundle is missing the primary execution script (scripts/generate.js), which is a significant red flag as the documentation (skill.md) explicitly states the tool accesses sensitive local OAuth credentials from 'auth-profiles.json'. It targets an internal Google sandbox endpoint (daily-cloudcode-pa.sandbox) and references non-public model versions ('Gemini 3 Pro'), making it impossible to verify if the credentials are being used legitimately or exfiltrated without the source code.
能力评估
Purpose & Capability
Name/description (Antigravity/Gemini image generation) aligns with required binary (node) and declared config path (auth.profiles). However the skill claims to call an internal Google sandbox endpoint (daily-cloudcode-pa.sandbox) — acceptable for an internal tool but unusual for a third‑party skill with unknown source and no homepage.
Instruction Scope
SKILL.md instructs running /home/ubuntu/clawd/skills/antigravity-image-gen/scripts/generate.js and explicitly states it will read local OAuth tokens from auth-profiles.json. But this skill bundle contains no code files (only SKILL.md and _meta.json) — the referenced script is missing. Instructions also encourage using sensitive local OAuth credentials to access an internal endpoint, which increases risk if the actual script behaves differently.
Install Mechanism
No install spec (instruction-only), so nothing is written to disk by the skill installer itself. This is low-risk from an install mechanism perspective.
Credentials
The only declared config requirement is auth.profiles, which matches the instructions' stated need to read OAuth tokens. Reading OAuth credentials is sensitive but can be proportionate for an image-generation client that authenticates to an API. Still, the skill requests access to local OAuth tokens but provides no code to review, so the sensitivity is elevated.
Persistence & Privilege
The skill does not set always:true and does not request persistent system-wide privileges. Autonomous invocation is enabled by default, which is normal but should be considered along with the other concerns.
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 clawhub-publish-146230 - 安装完成后,直接呼叫该 Skill 的名称或使用
/clawhub-publish-146230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
v2.0.0
Version 2.0.0 introduces a major update with new image generation capabilities.
- Now generates images using the internal Google Antigravity API (Gemini 3 Pro Image) for high-quality, native output.
- Removes the need for browser automation by connecting directly via OAuth credentials.
- Adds flexible command-line options for prompt, output path, and aspect ratio selection.
- Handles common error cases with clear troubleshooting guidance.
- Outputs images in a way that Clawdbot can automatically detect and display results.
元数据
常见问题
Clawhub Publish 146230 是什么?
Generate images using the internal Google Antigravity API (Gemini 3 Pro Image). High quality, native generation without browser automation. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 305 次。
如何安装 Clawhub Publish 146230?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install clawhub-publish-146230」即可一键安装,无需额外配置。
Clawhub Publish 146230 是免费的吗?
是的,Clawhub Publish 146230 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
Clawhub Publish 146230 支持哪些平台?
Clawhub Publish 146230 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 Clawhub Publish 146230?
由 gxw975(@gxw975)开发并维护,当前版本 v2.0.0。
推荐 Skills