Urgent.News

What's breaking now, across thousands of outlets.

AI

Sierra and Meta propose a standard for personal AI agents

Sierra and Meta announced the Personal Agent Protocol on October 6, 2026, a proposed open standard for how a shopper's AI agent should deal with a business. Sierra says the first spec, version 0.1, will arrive "later this month," according to Sierra's announcement . No spec exists yet. Today agents often act like a person clicking through a website, and the standard aims to replace that with a…

On October 6, 2026, Sierra and Meta introduced the Personal Agent Protocol (PAP) as an open standard for personal AI agents interacting with businesses. The first version, v0.1, is expected later this month. PAP aims to replace the current practice of agents acting like human users on websites with a set of clear rules. The protocol covers how the AI agent proves its affiliation, the customer's control, and the company's visibility.

It works by starting on the company's website, with the agent determining what the company offers. The customer can choose read-only or write access for account tasks, with their consent determining the level of control. Sessions use OAuth for sign-in and carry across channels. Businesses can choose to integrate with the protocol via web pages, APIs, or by allowing the company's personal agent to take over the conversation.

Major companies like Meta, Genesys, Shopify, Stripe, and Walmart are involved in developing and supporting PAP. The standard does not cover payments, future extensions include handling payments and push notifications. Developers building shopping, support, or commerce software should prepare for the standard's implementation within months, focusing on clean APIs, designing for guest and signed-in modes, and planning for multiple standards.

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

Also reported by 1 other outlet

Read the original at dev.to →

More in AI

Where should an AI agent's idempotency key come from?

Your agent's worker sends a payout request, and then the process gets killed before the response comes back. The supervisor restarts it.

  • Idempotency key should be generated during action planning, not execution.
  • Persist the key with task state before first attempt, in databases, queues, or workflows.

More from Thursday 8 October →