Simon Willison 在 Codex Desktop 中创建自定义宠物"Pedalican"

simonw/pedalican：https://github.com/simonw/pedalican。显然，当这些首次发布时：https://twitter.com/OpenAIDevs/status/2050301642717950166 我没有注意到，但今天我在 Codex Desktop 中意外激活了一个“宠物”——一个小型动画机器人，让人联想到 Clippy：https://en.wikipedia.org/wiki/Office_Assistant ——然后我发现你可以创建自己的宠物。

于是我创建了，现在我的桌面上有一只可爱的小鹈鹕骑着自行车，跳来跳去，给我更新我的 Codex 任务。

您的浏览器不支持 HTML5 视频。

这个过程中最有趣的事情是观察自定义宠物是如何被创建的。我告诉它我想要一个骑自行车的鹈鹕的自定义宠物，然后 GPT-5.6 Sol xhigh 完成了剩下的工作，使用 gpt-image-2：https://developers.openai.com/api/docs/models/gpt-image-2 多轮生成必要的精灵图素材。

我让它做了详细的记录：https://github.com/simonw/pedalican-pet/blob/main/notes-on-creating-a-pet.md 并记录所有中间步骤：https://github.com/simonw/pedalican-pet/tree/main/run。我的 GitHub 仓库包含每个生成的图像和组合精灵表，以及每个动画循环的 GIF，例如这个名为 waving.gif 的动画：https://github.com/simonw/pedalican-pet/blob/main/run/qa/previews/waving.gif

这个 GIF 是由单张图像编译而成：https://github.com/simonw/pedalican-pet/blob/main/run/api-generation/waving.png，由 gpt-image-2 生成，看起来是这样的：

而这张图像是通过执行这个提示：https://github.com/simonw/pedalican-pet/blob/main/run/prompts/rows/waving.md 对初始生成的角色参考图像进行创建的：https://github.com/simonw/pedalican-pet/blob/main/run/api-generation/base.png，这张参考图像是使用此提示创建的：https://github.com/simonw/pedalican-pet/blob/main/run/prompts/base-pet.md，其结构如下：

为 Codex 宠物 Pedalican 创建一个干净的全身参考精灵。

宠物身份：一只紧凑可爱的鹈鹕宝宝，圆圆的奶白色身体，柔软的珊瑚橙色嘴和脚，骑着一辆小小的天蓝色自行车 [...]

在一个完全平坦的纯品红 #FF00FF 色的色键背景上放置一个居中的单个姿势。保持整个宠物可见、紧凑、在192x208下易于辨认，并且容易动画化。 [...]

我一直在寻找利用图像生成创建简单游戏就绪精灵的方法，所以我花了一些时间研究这个机制，看看它是如何工作的。

关键的实现细节是开源的——特别是这两个技能，都采用 Apache 2.0 许可：

是的，GPT-5.6 Sol 确实想出了“Pedalican”这个名字。我很喜欢它！

这是 Simon Willison 于 2026 年 7 月 14 日发布的链接帖子：/2026/Jul/14/。

每月资助我 10 美元，可获得经过精选的当月最重要的大型语言模型发展的电子邮件摘要。

simonw/pedalican：https://github.com/simonw/pedalican . Clearly I wasn't paying attention when these were first announced：https://twitter.com/OpenAIDevs/status/2050301642717950166 back in May, but today I accidentally activated a "pet" in Codex Desktop - a little animated robot, reminiscent of Clippy：https://en.wikipedia.org/wiki/Office_Assistant - and then learned you can create your own.

So I did, and now I have a cute little pelican on a bicycle bouncing around my desktop giving me updates on my Codex tasks.

Your browser does not support HTML5 video.

The most interesting thing about this process was watching how the custom pet was created. I told it I wanted a custom pet that was a pelican riding a bicycle and GPT-5.6 Sol xhigh did the rest of the work, using several rounds with gpt-image-2：https://developers.openai.com/api/docs/models/gpt-image-2 to generate the necessary sprite assets.

I had it make extensive notes：https://github.com/simonw/pedalican-pet/blob/main/notes-on-creating-a-pet.md and record all of the intermediary steps：https://github.com/simonw/pedalican-pet/tree/main/run. My GItHub repo includes every generated image and combined sprite sheet, plus GIFs for each of the animation loops such as this one, called waving.gif：https://github.com/simonw/pedalican-pet/blob/main/run/qa/previews/waving.gif:

That GIF was compiled from a single image：https://github.com/simonw/pedalican-pet/blob/main/run/api-generation/waving.png generated by gpt-image-2 that looked like this:

And that image was created by executing this prompt：https://github.com/simonw/pedalican-pet/blob/main/run/prompts/rows/waving.md against the initial generated character reference image：https://github.com/simonw/pedalican-pet/blob/main/run/api-generation/base.png, which was created with this prompt：https://github.com/simonw/pedalican-pet/blob/main/run/prompts/base-pet.md, which has this structure:

Create one clean full-body reference sprite for Codex pet Pedalican.

Pet identity: A compact adorable baby pelican with a round cream-white body, soft coral-orange bill and feet, riding a tiny sky-blue bicycle [...]

Place a single centered pose on a perfectly flat pure magenta #FF00FF chroma-key background. Keep the full pet visible, compact, readable at 192x208, and easy to animate. [...]

I've been looking out for ways to use image generation to create simple game-ready sprites, so I spent some time digging into this mechanism to see how it works.

The key implementation details are open source - these two skills in particular, both Apache 2.0 licensed:

And yes, GPT-5.6 Sol did come up with the name "Pedalican". I like it!

This is a link post by Simon Willison, posted on 14th July 2026：/2026/Jul/14/.

Sponsor me for $10/month and get a curated email digest of the month's most important LLM developments.