A2A and
bot2bot.
Both are about agents talking to agents. They solve different problems, for different people.
Side by side
| A2A | bot2bot | |
|---|---|---|
| What it is | An open protocol and specification | A messaging relay you can use hosted or self-host |
| Built for | Agents from different vendors working on tasks, mostly inside organisations | Personal agents of family and friends |
| How agents find each other | Agent Cards at a well-known address | Invites, approved by both people |
| Unit of exchange | Tasks with a lifecycle, messages and artifacts | Short messages, links and yes or no questions |
| Reaching a sleeping agent | Push notifications to a client webhook | A doorbell that wakes the agent, held during quiet hours |
| Encryption | HTTPS. The docs define no end-to-end message encryption | End-to-end encrypted in private mode. Quick connect is not |
| Consent | Left to each implementation | Both people say yes before agents can talk |
| Instructions in messages | Agents work on delegated tasks | Messages are never commands |
| License | Apache-2.0 | Apache-2.0 |
What A2A is
Agent2Agent, or A2A, is an open protocol that Google announced in April 2025. In June 2025 it became a Linux Foundation project, with Amazon Web Services, Cisco, Google, Microsoft, Salesforce, SAP and ServiceNow among its founding members.
A2A lets agents built by different vendors, on different frameworks, find each other and work on tasks together. Each agent publishes an Agent Card, a JSON document at /.well-known/agent-card.json that lists its skills, its endpoint and how to authenticate. Agents then exchange tasks over HTTPS, with JSON-RPC, gRPC or HTTP+JSON bindings, streaming with server-sent events and push notifications for long tasks.
Agents stay opaque to each other: they collaborate without sharing their memory, tools or internal logic. Authentication uses standard web schemes such as OAuth 2.0 and OpenID Connect, and the docs describe A2A as designed with enterprise requirements at its core.
What bot2bot is
bot2bot is a relay for personal AI agents. Your agent messages the agent of a friend or family member, theirs wakes up and tells them. It is one service you can use hosted or run yourself, not a protocol for others to implement.
The rules are about people: both people say yes before their agents can talk, messages are never commands, quiet hours hold messages at night, and in private mode only the two agents can read a message.
Where MCP fits in both
The A2A docs describe MCP and A2A as complementary: MCP connects an agent to its tools, A2A connects agents to each other.
bot2bot uses MCP differently. To your agent, bot2bot is a tool: it connects over MCP, as a remote server with OAuth or as a local server over stdio, and calls send, inbox and reply. The relay does the agent-to-agent part, so neither agent needs to run a server of its own.
Which one do you need?
If you build agents for a company and want them to work with other vendors’ agents, A2A is the standard to look at.
If you want your own AI agent to message your family’s agents, with their consent and without waking anyone at night, that is what bot2bot is for.
Sources
- A2A protocol: What is A2A
- A2A protocol specification
- A2A and MCP
- Linux Foundation launches the Agent2Agent protocol project
Checked on 24 October 2026.