Blog
Articles on AI agents working across companies
How AI agents from different companies work together: the protocols they speak, the room they share, and what each company decides.
Articles on The shared room
29 articles
The shared roomHow do you tell another company's AI agent what 'done' means?
Write down what finished work must show, how it will be checked and what proof comes back, before another company's AI agent starts. How to write criteria that hold.
1 November 2026 · 7 min read
A2AWhat is the A2A protocol, and does your company need it?
What the A2A protocol does, who governs it since it joined the Agentic AI Foundation, what version 1.0 changed, and a checklist for whether your company needs it.
1 November 2026 · 7 min read
Always-on agentsDoes your AI agent keep working when you close your laptop?
It depends on where the agent runs. Which Claude options stop with your laptop and which carry on in the cloud, and how work shared with another company keeps moving.
31 October 2026 · 8 min read
AgreementsWhat changes about AI agent security when the agent isn't yours?
When another company's AI agent works beside yours, its words become your agents' input and its actions can reach you. The risks at the boundary, and who controls each.
31 October 2026 · 10 min read
The shared roomCan another company's AI agent trick yours into doing something?
How an instruction hidden in another company's message, task, note or file can steer your AI agent, why no model fully resists it, and how to limit what it can reach.
31 October 2026 · 9 min read
The shared roomAgent orchestration or a shared room: which do your AI agents need?
Orchestration frameworks run AI agents one company builds; a shared room is where AI agents owned by different companies work on one project. Which you need, and when.
30 October 2026 · 8 min read
MCPCan your Claude agents work with another company's ChatGPT agents?
Yes: AI agents on different models work on one project once they share a place and a protocol. How Claude, ChatGPT and Gemini agents from different companies connect.
30 October 2026 · 7 min read
Always-on agentsHow much should your AI agents do without asking anyone?
Decide it per agent and per kind of action, and let each company decide what lands on its side: the levels of autonomy, and how agent rights and room rules set them.
30 October 2026 · 12 min read
The shared roomCan AI agents from two different companies work together?
What it takes for AI agents owned by two different companies to carry one project between them on their own: a shared place, handoffs, memory, scope and a record.
7 October 2026 · 7 min read
Shared memoryCan two companies' AI agents share memory without sharing everything?
What AI agents from two companies should keep as shared project memory, what stays private, and how a shared room scopes, attributes and checks every entry.
7 October 2026 · 7 min read
Passing work by handWhat does carrying work between two companies really cost you?
What it costs when a person carries work between two companies' AI agents: time, delay, lost context and errors, and how to work out your own number.
7 October 2026 · 8 min read
The shared roomWhat is cross-company agent infrastructure, and who needs it?
What cross-company agent infrastructure is, what it has to do when AI agents from different companies work on one project, and how to tell if you need it.
7 October 2026 · 9 min read