Sato Hub
Standard

A2A: agents delegating to agents

As of 2026-10-15, 14 listed tools say they support Agent2Agent; highest Sato Scores: Bankr, Agent Payments Protocol (AP2) and Lucid Agents.data: JSON · CSV

  • Implementations14
  • Sato Scored14
  • Deploy specs9
  • Updated2026-10-15

A2A covers the case MCP doesn't: not an agent calling a tool, but an agent handing work to another agent. Each participant publishes a card describing what it can do and how to reach it, and an orchestrator reads those cards to decide where to route a task.

It's the newest of the standards tracked here and the least settled. The listings below implement some part of it; read each one for which part, because coverage varies a great deal.

A2A: 14 listings, 10 open source and 4 with a commit or release in the last 30 days; top Sato Scores: Bankr (87), Agent Payments Protocol (AP2) (72), Lucid Agents (71), as of 2026-10-15.

API / SDK

3

Payment Rail

3

MCP

2

Agent Framework

1

Agent Launchpad

1

Developer Tool

1

Security Tool

1

Skill Repo

1

Wallet Infrastructure

1

Other standards

Reference

QuestionsShowHide

Questions people ask about A2A: agents delegating to agents

How many projects implement Agent2Agent?
14 live listings declare it as of 2026-10-15, spread across api / sdk 3, payment rail 3, mcp 2, agent framework 1. A declaration is recorded from the project's own materials — it is the floor of adoption, not a measurement of it.
Does declaring Agent2Agent mean it answers?
No, and the gap is the point. Declaring costs nothing: a tag, a manifest line, a sentence in a README. Whether an endpoint actually responds is measured separately by daily checks, published per listing, and never folded into this count.
Where is Agent2Agent defined?
In the standard's own documents, not here: A2A protocol at https://a2a-protocol.org. Sato Hub records which listings implement it and what our checks observed; it does not author the standard and does not certify conformance.
Are these implementations verified?
Not by implementing the standard. 14 of the 14 carry a Sato Score — a 0–100 reading of how open, active and verifiable a project is — and 9 publish an install manifest. Neither is an audit, a security review, or a conformance certificate.
Which of these are open source?
10 of 14: Agent Payments Protocol (AP2), Lucid Agents, Universal Crypto MCP, Nevermined, Emblem Agent Skills, PRXVT, Agently, Universal Commerce Protocol (UCP), Agent0 SDK, create-8004-agent.
Which are actively maintained?
4 had a commit or release in the last 30 days: Bankr, Universal Crypto MCP, Nevermined, Hashnet MCP Server. Posts on X do not count. Activity is a recency signal, not a quality judgment.
Which have an earned verification on Sato Hub?
8: Bankr, Agent Payments Protocol (AP2), Lucid Agents, Universal Crypto MCP, Emblem Agent Skills, PRXVT, Agently, Universal Commerce Protocol (UCP). An earned check is named as what was checked (install reproduced, endpoint probed, on-chain identity read, reviewed against evidence); none is a safety claim.
Which run as MCP servers?
7: Agent Payments Protocol (AP2), Lucid Agents, Universal Crypto MCP, Emblem Agent Skills, Hashnet MCP Server, Agently, Universal Commerce Protocol (UCP).
How is this list ranked?
A2A is derived from Sato Hub's live records by a stated rule and ordered by Sato Score, a 0–100 measure of how open, active and verifiable each project is. It is not a quality ranking and declares no winner.
SourcesShowHide
Cite this pageShowHide

Cite this page

Sato Hub. "A2A: agents delegating to agents." Sato Hub, updated 2026-10-15, accessed 2026-10-15. https://satohub.ai/standards/a2a

Data last refreshed 2026-10-15; this page is rebuilt daily. Citations carry the date so a reader can tell which snapshot a claim came from. Catalog data is licensed CC-BY-4.0.

Join the Sato Hub Briefing

Sign up for the Briefing: the agents, tools, and infrastructure that actually shipped, and why they matter.