Port a computer-use tool into another harness
They had Gemini rework Codex’s computer-use tool for Antigravity and finished the port in under 10 minutes.
- What they did
- They gave Gemini access to Codex’s folder on their PC and asked it to make the computer-use tool work inside Antigravity. Gemini turned the Codex binary into a bridge with no extra dependencies, usable from both the command line and an MCP stdio server. It also dealt with Windows quirks like sandboxed window access, auto-approving newly opened apps, bringing back a hidden pointer, and falling back to screenshots for Chrome or minimized windows.
- How it went
- The poster says the adapted tool reached feature parity with Codex’s version. They used it successfully for autonomous Telegram Web messaging on a setup with two monitors and for complete Antigravity interface control. Early trouble included an invisible cursor and browser-capture failures.
- Tips
- Ports like this can hinge on low-level desktop behavior, especially on Windows. If you try it, ask the agent to pause before any destructive or final action so you can review the changes first.
Try it yourself with Gemini
Take the files in [folder path] for [source tool] and modify them so the same tool runs inside [target harness or environment]. Keep the original behavior as much as you can, and make the finished version available in [output folder]. Add a short install guide and one basic test I can run to verify it works. Stop and ask me before you do anything irreversible, including deleting files, sending data, making payments, or submitting anything. When you’re done, list the changes you made, the parts that were not portable, and any OS-specific problems you had to solve.
Fill in the [brackets]. Agents can make mistakes, so review anything before it is sent, paid or deleted.
Get 3 agent uses like this every morning
With the day's top AI news. Free, and one click to unsubscribe.
Related agent uses
Run agent coding jobs in a sealed local box
A developer used Codex to run an almost fully self-contained build-and-deploy workflow on a locked-down spare machine.
Create and launch a BaZi reading agent
A developer used coding agents to build a BaZi reader that calculates charts, writes interpretations, and handles language detection and USDT payments.
Review a pull request with staged sub-agents
Use parallel checks and a validation pass to review a branch or PR before you decide what to fix.
Contain agent runs and clear old temp files
A developer limited each coding run to a separate cgroup and added a recurring cleanup for forgotten temp files.
Use an agent to patch a stubborn app bug
They pointed a coding agent at two annoying software behaviors and had it debug one and add a self-improvement feature to the other.
Run several coding agents from one interface
A developer uses T3 Code to manage multiple coding agents on personal machines, a VPS, and a separate work setup for code changes and writeups.