Urgent.News

What's breaking now, across thousands of outlets.

AI

How I built a managed ChatGPT browser for my desktop AI companion: no pairing link, no borrowed tab

How I built a managed ChatGPT browser for my desktop AI companion: no pairing link, no borrowed tab The worst setup flow I shipped was the one that worked. To connect my desktop companion to ChatGPT, you copied a pairing link out of the app, pasted it into your browser, and hoped the two ends agreed about who they were talking to. It worked — and every user who did it had to understand my…

A new managed ChatGPT browser has been created for desktop AI companions, eliminating the need for a pairing link or borrowing a tab. Instead of users copying a pairing link and pasting it into their browser, the app now owns the browser end to end. In this managed browser, the app discovers the browser automatically, and the connection between them is secure.

Two modes are available: builtin and Chrome. In builtin mode, the ANKITA window opens inside the app itself, and the user's sign-in is saved. The Chrome mode launches Chrome if it is absent, automatically connecting and opening the right profile. A stable host ID is minted once with randomUUID() and stored in the app's settings, ensuring that the browser is linked to the correct app.

The app's sign-in is protected, only allowing navigation to a small allowlist of login hosts. Saved sign-in never launches anything on startup, and the user must initiate the process each time they want to use the companion. The app also refuses to switch browser or teammate while a reply is pending, ensuring that the user is aware of the connection before proceeding.

While this managed browser offers many advantages, it does have some limitations. Discovery only works on loopback, meaning there is no fallback if the companion cannot reach the app on the same machine. The Chrome mode depends on the companion extension staying installed in exactly one profile, and the builtin window is a single-purpose window. Additionally, the design is built around the subtle bug that a loopback port is not an identity, ensuring that multiple apps cannot claim the same connection.

Overall, this new managed ChatGPT browser for desktop AI companions offers a more secure and user-friendly experience, eliminating the need for a pairing link and reducing the risk of errors caused by hand-carried credentials.

Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

This story

This is one outlet's version. Read the fullest account.

Read the original at dev.to →

More in AI

What smart-contract audits taught me about evaluating LLMs

Before I worked on LLM systems, I spent a lot of my time in smart-contract audit contests. I entered more than twenty of them, on platforms like Sherlock, Code4rena and Immunefi.

  • Author's experience in smart contract audit contests
  • Transitioned to building AI production systems
  • Emphasized worst-case scenario evaluation for LLMs

Anthropic IPO: Why There Could Be More Than One AI Winner

Instead of attempting to guess whether Anthropic, OpenAI, or one of the other AI competitors will dominate, investors should consider a different possibility.

  • Anthropic's IPO in late 2026 will offer retail investors ownership in Claude's AI.
  • Contrary to "winner-take-all" view, multiple players can succeed in large markets.
  • AI market projected at $1.9 trillion by 2032, allowing both Anthropic and OpenAI to thrive.

More from Saturday 31 October →