Comparisons
What each protocol and approach is for, where they overlap, and when you would use both.
A2A vs ACP: how IBM's agent protocol merged into A2A
IBM's Agent Communication Protocol merged into A2A in August 2025. What ACP was, how its concepts map to A2A, and what ACP users were told to do.
Updated
A2A vs Agora: one fixed protocol or negotiated ones
A2A gives every agent the same task protocol. Agora, a 2024 research proposal, lets LLM agents negotiate their own formats. What each is for, and status.
Updated
A2A vs ANP: two designs for agents on the open web
A2A delegates tasks to agents that publish Agent Cards. ANP gives each agent a DID and a crawlable description. How they differ, and where ANP 1.2 stands.
Updated
A2A vs GraphQL: who decides the shape of the answer
GraphQL lets a client ask for exactly the data it wants from a typed schema. A2A lets an agent hand a goal to another agent. How the two differ and combine.
Updated
A2A vs IVR: phone menus for callers, tasks for agents
An IVR routes phone callers through prompts with keypad or speech input. A2A lets agents send a business structured tasks. How they differ and coexist.
Updated
A2A vs OpenAPI: handing off a goal or calling a described API
OpenAPI describes an HTTP API's operations so callers can use them exactly. A2A lets an agent hand a goal to another agent and track the task. When each fits.
Updated
A2A vs REST APIs: delegating a task or operating on a resource
A REST API gives callers exact operations on a service's resources. A2A lets one agent hand a goal to another and track the task. How they differ and fit.
Updated
A2A vs webhooks: push notifications with a task around them
A2A push notifications are webhooks scoped to one task, with a fixed payload and receiver-chosen credentials. How they differ from plain webhooks.
Updated
Agent Card vs llms.txt: a connection contract and a reading list
An Agent Card tells client agents how to connect to an A2A agent. llms.txt, a community proposal, points models to a site's best content. Why publish both.
Updated
Agent Card vs OpenAPI document: what each describes and who reads it
An Agent Card describes an A2A agent: identity, endpoints, skills and auth. An OpenAPI document describes an HTTP API's operations. Field by field.
Updated
Agent-to-agent vs chatbots: a conversation for people, a contract for software
A chatbot holds a conversation with a person in a chat channel. Agent-to-agent protocols like A2A let software agents exchange structured tasks. When each fits.
Updated
Agent-to-agent vs RPA: driving the screen or asking the agent
RPA robots operate applications through their user interface. A2A lets agents hand each other tasks directly. How the two differ and combine.
Updated
AP2 vs Agentic Commerce Protocol: authorization evidence and checkout
AP2 v0.2 proves what a user let an agent buy and pay for. OpenAI and Stripe's Agentic Commerce Protocol runs the checkout. How they differ, with current status.
Updated
Google ADK vs LangGraph for A2A: what each supports as of September 2026
Google's ADK can serve and call A2A agents, marked experimental. LangGraph's Agent Server serves A2A 1.0 over JSON-RPC. Versions, features and gaps.
Updated
JSON-RPC vs gRPC for A2A: two bindings for one protocol
A2A's JSON-RPC binding sends JSON over HTTP with SSE streaming. Its gRPC binding sends Protocol Buffers over HTTP/2. How the two compare and when to offer both.
Updated
LangGraph vs CrewAI for A2A: what each supports as of September 2026
LangGraph's Agent Server exposes graphs as A2A 1.0 JSON-RPC endpoints. CrewAI agents can call and serve A2A at protocol 0.3. Versions, features and gaps.
Updated
llms.txt vs robots.txt: a reading list and a crawl policy
robots.txt (RFC 9309) tells crawlers which paths they may fetch. llms.txt, a community proposal, points models to a site's best content. How they differ.
Updated
MCP vs function calling: where tools come from and who runs them
Function calling is how a model asks your app to run a tool. MCP is how your app finds and calls tools on a server. How the two layers differ and fit together.
Updated
OAuth scopes vs mandates: what a token allows and what a person approved
OAuth scopes and RFC 9396 authorization details limit what a token can do. Mandates such as AP2's prove what a person approved an agent to do. How they compare.
Updated
Signed Agent Cards vs mTLS: what each one proves
A signed Agent Card proves a card is intact and came from a key holder. Mutual TLS proves who is on each end of a connection. Where each applies.
Updated
SSE streaming vs push notifications in A2A: hold a connection or register a webhook
A2A streaming sends task events over an open SSE connection. Push notifications POST them to a client's webhook. How they differ, and when to poll.
Updated
Voice AI vs A2A for customer service: serving callers and serving agents
Voice AI agents answer people on the phone in natural speech. A2A accepts structured tasks from customers' agents. How they differ and share a back end.
Updated
Web Bot Auth vs CAPTCHA: naming the bot or testing for a human
A CAPTCHA tries to tell humans from bots and keep bots out. Web Bot Auth lets bots and agents sign requests so a site knows who sent them. How they compare.
Updated