← 返回 Skills 市场
VPick AI Image Generator
作者
snoopyrain
· GitHub ↗
· v1.0.2
· MIT-0
140
总下载
0
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install vpick-image-generator
功能描述
Multi-model AI image generation on a visual canvas. Supports Midjourney (relaxed/fast/turbo, v7.0, 4-image grid), Grok Imagine (6 images per call, auto I2I),...
安全使用建议
This skill is coherent for remote image generation but relies on an external VPick connector token (the MCP link). Treat that MCP link like a password: only paste it into connectors you trust and revoke it if compromised. Understand that your prompts and any uploaded reference images will be forwarded to VPick (vpick.10xboost.org) and routed to third-party model providers, and that generated assets will be stored in Google Cloud Storage under your VPick account. Before installing or using, verify the VPick service is legitimate (check the homepage, privacy policy, and billing terms), avoid uploading highly sensitive images, and test with non-sensitive prompts/credits first. If you need tighter privacy or on-premise generation, this skill may not be appropriate.
功能分析
Type: OpenClaw Skill
Name: vpick-image-generator
Version: 1.0.2
The skill bundle consists of metadata and documentation (SKILL.md) for an image generation service called VPick. It provides instructions for an AI agent to interact with various image models (Midjourney, Grok, etc.) via a set of tools like `run_image_generator` and `add_node`. The documentation is transparent about data handling, noting that requests are sent to `vpick.10xboost.org` and that authentication is handled via an MCP link token. No malicious code, exfiltration patterns, or harmful prompt injections were found.
能力评估
Purpose & Capability
The name/description (multi-model image generation) match the SKILL.md instructions (create image nodes, set prompts, upload references, run generator). No unrelated credentials, binaries, or installs are requested.
Instruction Scope
Instructions are limited to canvas operations (get_canvas, add_node, upload_image, run_image_generator, etc.) and describe sending prompts/images to VPick. They do not instruct the agent to read local files, environment variables, or other system state beyond the canvas/connector context.
Install Mechanism
No install spec or code files are present. This instruction-only skill does not download or write code to disk, which minimizes installation risk.
Credentials
The skill declares no required environment variables or credentials. It relies on an external MCP link (containing an embedded token) entered by the user as a connector — this is proportional to the stated design and explicitly documented in the SKILL.md.
Persistence & Privilege
always is false and the skill is user-invocable with normal autonomous invocation allowed. The skill does not request persistent system-wide changes or permissions beyond using the user-provided MCP connector/token in the agent's connector settings.
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 vpick-image-generator - 安装完成后，直接呼叫该 Skill 的名称或使用
/vpick-image-generator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
v1.0.2
- Added detailed Security & Data Handling section, covering authentication, data flow, storage, and billing practices.
- Clarified that the MCP link contains the only required authentication token, with no separate API keys or secrets needed.
- Explained how prompts and reference images are handled and routed to third-party model providers.
- No changes to core workflow or model support.
- Documentation now more clearly describes privacy, security, and storage of generated images.
v1.0.0
Initial release: multi-model AI image generation on a visual canvas with advanced features.
- Supports Midjourney (v7.0, relaxed/fast/turbo), Grok Imagine, nano-banana-2, and Seedream 5.0 models.
- Features include text-to-image, image-to-image, style transfer, multi-reference input, batch generation, and aspect ratio control.
- Allows adding, connecting, and managing nodes for complex image generation workflows.
- Integrates reference images using "@NodeLabel" for guided generation and style transfer.
- Includes detailed setup, error handling, and workflow instructions for easy onboarding and use.
元数据
常见问题
VPick AI Image Generator 是什么？
Multi-model AI image generation on a visual canvas. Supports Midjourney (relaxed/fast/turbo, v7.0, 4-image grid), Grok Imagine (6 images per call, auto I2I),... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 140 次。
如何安装 VPick AI Image Generator？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install vpick-image-generator」即可一键安装，无需额外配置。
VPick AI Image Generator 是免费的吗？
是的，VPick AI Image Generator 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
VPick AI Image Generator 支持哪些平台？
VPick AI Image Generator 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 VPick AI Image Generator？
由 snoopyrain（@snoopyrain）开发并维护，当前版本 v1.0.2。
推荐 Skills