← 返回 Skills 市场
Raphael Ai Image
作者
CharlieDaQin
· GitHub ↗
· v0.2.0
· MIT-0
152
总下载
0
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install raphael-ai-image
功能描述
通过输入文字提示词，使用 Raphael AI 免费生成高质量无水印图片，支持24种视觉风格和电商优化场景。
安全使用建议
This skill appears to do exactly what it says: it automates a browser to use raphaelai.org, downloads generated images from cdn.raphaelai.org, saves them under ~/.openclaw/workspace/media/, and posts them to Discord. Before installing, confirm you trust the external site (https://raphaelai.org) and the agent's Discord integration because generated images will be downloaded and transmitted to a channel. There are no requested API keys or other secrets. If you do not want the agent to access the network, write files under your home directory, or post to Discord automatically, do not enable this skill. (Minor metadata/version fields in _meta.json differ from registry metadata but this is likely a non-security bookkeeping issue.)
功能分析
Type: OpenClaw Skill
Name: raphael-ai-image
Version: 0.2.0
The skill bundle is a legitimate automation tool for generating images using the Raphael AI website. The SKILL.md file provides clear instructions for browser automation, and the scripts/generate_image.py file serves only as a documentation holder for styles and prompts without executing any harmful code. No evidence of data exfiltration, malicious execution, or prompt injection was found.
能力评估
Purpose & Capability
Name/description promise (generate watermark-free Raphael AI images) matches the SKILL.md and the included documentation script which describe using OpenClaw Browser to navigate raphaelai.org, extract image URLs from cdn.raphaelai.org, download, save, and post to Discord. There are no unrelated credentials, binaries, or config paths required beyond the agent's normal browser/network capabilities.
Instruction Scope
Instructions explicitly direct the agent to open the external site, interact with the DOM, extract a CDN URL, curl the image, save to ~/.openclaw/workspace/media/, and send via Discord. Those steps are appropriate for the stated task, but they do involve network access, writing files to the agent workspace, and use of the agent's Discord channel integration — users should be aware images will be downloaded and posted externally.
Install Mechanism
No install spec is provided (instruction-only plus a documentation-only Python file). Nothing will be downloaded or executed at install time by the skill itself.
Credentials
The skill requires no environment variables, no keys, and no special config paths. The only implicit requirement is the agent's ability to access the network, run browser automation, and post to Discord (typical platform capabilities).
Persistence & Privilege
always is false, and the skill does not request persistent elevated privileges or modify other skills' configs. It writes outputs to the agent workspace path (~/.openclaw/workspace/media/) which is consistent with normal media handling.
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 raphael-ai-image - 安装完成后，直接呼叫该 Skill 的名称或使用
/raphael-ai-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
v0.2.0
- 简化并明确了技能的定位、调用流程与角色分工（主 Agent/请求方）。
- 提升文档结构，突出快速参考与常用模板，便于用户和 Agent 快速上手。
- 优化提示词和风格关键词参考，更聚焦实用表达。
- 汇总 24 种视觉风格，归纳适用场景，精炼说明。
- 精简技术说明和注意事项，强化“免注册、完全免费、可商用”等核心卖点。
v0.1.0
- 首发版本：通过 OpenClaw 自动操控 raphaelai.org，实现免费、无限、无水印的 AI 图片生成功能
- 支持输入任意提示词，生成高质量图片，适合电商、社媒、Logo、概念艺术等多场景
- 内置 24 种视觉风格，覆盖照片、动漫、电影感、像素、油画、赛博朋克等
- 图片可商用，自动优化提示词，生成 512x512 或更高分辨率无水印图片
- 操作简单，可全自动或手动选择风格，无需注册或 API Key
元数据
常见问题
Raphael Ai Image 是什么？
通过输入文字提示词，使用 Raphael AI 免费生成高质量无水印图片，支持24种视觉风格和电商优化场景。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 152 次。
如何安装 Raphael Ai Image？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install raphael-ai-image」即可一键安装，无需额外配置。
Raphael Ai Image 是免费的吗？
是的，Raphael Ai Image 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Raphael Ai Image 支持哪些平台？
Raphael Ai Image 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 Raphael Ai Image？
由 CharlieDaQin（@charliedaqin）开发并维护，当前版本 v0.2.0。
推荐 Skills