Agent-to-agent protocols

Work passes between agents and people without losing context.

Facts

Lane: Agents and organisations
Wave: 9 of 12 (build order, not a date)
Needs: 1 boxes all the way back through hard links
Unlocks: 4 boxes downstream through hard links
Leverage: 0.2 points · stands in the way of 3 of 53 stages · rank 33 of 57
Cluster: Trust plumbing and the agent economy

Evidence as of October 2026

A2A reached version 1.0 with more than 150 supporting organisations by April 2026. Multi-agent systems still fail often when agents misread each other.

Around it

Needs · 1

Agent identity and permissionscapability · Emerging

Unlocks · 5

Intent interfacescapability · Speculative
Agent-native organisationsstage · Emerging · helps, not required
Agent paymentscapability · Emerging · helps, not required
Multi-agent coordinationcapability · Emerging · helps, not required
Misbehaviour cascades in agent swarmsbrake · Emerging · helps, not required

Openings researchers' judgement, October 2026 scouting

Startup room 33/100 · rank 95 of 124 Research room 28/100 · rank 119 of 124

Crowding4 Lab overlap5 Small-team fit2 Buyer today2 Research openness2

Depth: light
Players: Google A2A and MCP under Linux Foundation Agentic AI Foundation
Wedge: No wedge in protocols; cross-organisation authorisation sits in identity and audit. none

01Axios: A2A to Agentic AI Foundation