Make Coding Less Lonely: A Step‑by‑Step Guide to Enable and Customize Codex’s Desktop Pet
This article walks through enabling Codex’s new desktop‑pet feature, selecting a built‑in pet, installing the Hatch Pet Skill from GitHub, creating a custom pet by describing or uploading an image, and using a generic cyber‑pet tool for other AI coding assistants.
Enable built‑in desktop pet in Codex
Open Codex Settings (lower‑left button), select Appearance , scroll to the bottom, expand Pet . Eight default pets are listed. Choose one and click the top‑right Wake Pet button to place the pet on the desktop. The pet remains visible and shows status notifications when a task makes progress.
Create a custom pet with Hatch Pet Skill
Install the skill by sending the command:
安装 https://github.com/openai/skills/tree/main/skills/.curated/hatch-pet 这个 SkillThen type hatch, select the installed Hatch Pet skill, and either describe the desired appearance or upload a reference image. Codex generates a main visual, then produces multiple pose images (failure, jump, crawl, wait, run, review, etc.). After roughly ten visual states are generated, Codex assembles and composites them into the final custom pet.
Return to the Appearance → Pet page, where the new pet appears alongside the built‑in options; select it to activate.
Technical details of the generation process
Codex dispatches several sub‑agents that run in parallel to create each pose.
Each pose image corresponds to a task state (e.g., “failure” shows a drooping lobster, “jump” shows a cute upward motion).
When all pose images are ready, Codex composites them into an animated sprite used by the desktop pet.
Alternative generic desktop‑pet tool
For users of OpenCode, Claude Code, or other AI coding assistants, the “Cyber Pet” tool Confirmo can be used. Download the appropriate installer from the GitHub releases page:
https://github.com/yetone/confirmo-releases
Select the installer matching the operating system and follow the same activation steps.
Signed-in readers can open the original source through BestHub's protected redirect.
This article has been distilled and summarized from source material, then republished for learning and reference. If you believe it infringes your rights, please contactand we will review it promptly.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.
