Agent-to-Agent Protocol

Agent-to-Agent Protocol, or A2A, is a protocol for remote agents to communicate with other agents or connect directly to frontends as part of distributed agent systems.

Key points

  • The n8n cluster noted Google's A2A launch as part of the broader future trajectory for agentic workflows, alongside multi-agent systems and long-running agent harnesses [src-005].
  • Google's A2UI post says A2A 1.0 has officially launched and serves as a robust transport for remote agents communicating with other agents or directly connecting agents to frontends [src-038].
  • A2A appears in the A2UI transport layer: A2UI can run over MCP, WebSockets, REST, AG-UI, A2A, or other transports [src-038].
  • In the generative UI stack, A2A is transport/coordination, while A2UI defines the UI the user touches [src-038].
  • This makes A2A complementary to Model Context Protocol rather than a direct replacement: both sit around agents, but they standardize different parts of agent communication and integration [src-038].
  • Google Cloud Next '26 operationalizes A2A inside Gemini Enterprise Agent Platform through Agent-to-Agent Orchestration and Agent Gateway support for agentic protocols including MCP and A2A [src-044].
  • Fmind's A2A videos add a practitioner-friendly distinction: A2A is the coordination protocol for agents talking to agents, while MCP is the tool/context protocol for agents talking to external capabilities [src-078].
  • The "Bleeding Agent" format frames A2A as part of a larger agent infrastructure stack that also needs skills, tool integrations, security boundaries, observability, and deployable coding discipline [src-078].

Related entities

Related concepts

2026-06-22 update

  • Google's A2A anniversary/update piece frames A2A as a collaboration protocol for secure task delegation and context sharing among agents [src-127].

2026-06-24 update

  • Google's cross-language ADK example uses A2A to connect a Python extraction agent and a Go validation agent without forcing teams to rewrite services into one language [src-145].

Source references

  • [src-005] Nate Herk cluster — Nate Herk — n8n cluster (18 videos)
  • [src-038] Google A2UI Team — "A2UI v0.9: The New Standard for Portable, Framework-Agnostic Generative UI" (2026-04-17)
  • [src-044] Thomas Kurian — "Welcome to Google Cloud Next '26" (2026-04-22)
  • [src-078] Mederic Hurier (Fmind) channel transcript cluster (62 saved transcripts, 2024-11-26 to 2026-05-14)
  • [src-127] Google Developers Blog – "How A2A is Building a World of Collaborative Agents" (2026-06-18)
  • [src-145] Shubham Saboo and Eric Dong – "Build Cross-Language Multi-Agent Team with Google's Agent Development Kit and A2A" (2026-06-22)

2026-08-17 protocol-tour update

  • Mete Atamel's NDC talk packages MCP, A2A, AG-UI and A2UI as complementary protocol layers. AI Watch retains the title/channel signal, but no transcript was initially available from the VPS and detailed protocol claims require caption verification [src-301].
  • [src-301] NDC Conferences — “Tour of Agent Protocols: MCP, A2A, AG-UI, A2UI” (verified 2026-08-17)

Historical recovered snapshot — 2026-08-18

Preserved from ai-watch-src329-335-okf-replacement-20260818t150256z. Documentary recovery on 2026-09-09; claims below retain their historical source scope and were not freshly verified. Current synthesis above remains authoritative.

  • The n8n cluster noted Google's A2A launch as part of the broader future trajectory for agentic workflows, alongside multi-agent systems and long-running agent harnesses [src-012].
  • Google's A2UI post says A2A 1.0 has officially launched and serves as a robust transport for remote agents communicating with other agents or directly connecting agents to frontends [src-038].
  • A2A appears in the A2UI transport layer: A2UI can run over MCP, WebSockets, REST, AG-UI, A2A, or other transports [src-038].
  • In the generative UI stack, A2A is transport/coordination, while A2UI defines the UI the user touches [src-038].
  • This makes A2A complementary to Model Context Protocol rather than a direct replacement: both sit around agents, but they standardize different parts of agent communication and integration [src-038].
  • Google Cloud Next '26 operationalizes A2A inside Gemini Enterprise Agent Platform through Agent-to-Agent Orchestration and Agent Gateway support for agentic protocols including MCP and A2A [src-044].
  • Fmind's A2A videos add a practitioner-friendly distinction: A2A is the coordination protocol for agents talking to agents, while MCP is the tool/context protocol for agents talking to external capabilities [src-078].
  • The "Bleeding Agent" format frames A2A as part of a larger agent infrastructure stack that also needs skills, tool integrations, security boundaries, observability, and deployable coding discipline [src-078].
  • [src-012] Nate Herk cluster — Nate Herk — n8n cluster (18 videos)
  • [src-038] Google A2UI Team — "A2UI v0.9: The New Standard for Portable, Framework-Agnostic Generative UI" (2026-04-17)
  • [src-044] Thomas Kurian — "Welcome to Google Cloud Next '26" (2026-04-22)
  • [src-078] Mederic Hurier (Fmind) channel transcript cluster (62 saved transcripts, 2024-11-26 to 2026-05-14)
  • [src-127] Google Developers Blog – "How A2A is Building a World of Collaborative Agents" (2026-06-18)
  • [src-145] Shubham Saboo and Eric Dong – "Build Cross-Language Multi-Agent Team with Google's Agent Development Kit and A2A" (2026-06-22)

2026-08-17 protocol-tour update

  • AAIF publicly announced A2A as a Growth Stage hosted project after Technical Committee approval on 15 July and Governing Board approval on 4 August. This adds AAIF governance to an existing Linux Foundation project; it is not a fresh protocol launch or donation [src-331].
  • A2A v1.0 was already stable before the transition, and its Apache-2.0 specification covers discovery, agent cards, task delegation, status and result exchange. Neutral hosting does not itself prove universal interoperability, security or adoption [src-331].
  • [src-331] Agentic AI Foundation — “A2A joins AAIF’s open agentic stack” (2026-08-17)

Robin Cartier perspective

This page is part of Robin Cartier's working AI knowledge graph: a practical research layer for production AI, recommendation systems, experimentation, GEO, and agentic web readiness.

The useful next step is to connect this concept back to applied product leadership and operating models.

Recommended next

Keep reading from this thread

From 500 indexed pages and articles.

  1. Wiki concept A2UI A framework-agnostic standard for declaring AI-generated UI intent so agents can drive client interfaces using the client's existing component catalog. Related by a2ui
  2. Wiki concept MCP Apps Custom, sandboxed application surfaces that can be exposed through the Model Context Protocol and embedded into agentic user experiences when declarative Related by a2ui
  3. Wiki concept Agentic AI Mesh QuantumBlack / McKinsey's architecture pattern for scaling agents across an enterprise without turning every workflow into a bespoke, unmanaged Related by a2a