{
  "slug": "a2a-authentication-schemes",
  "title": "A2A Authentication Schemes: Which to Declare in Your Agent Card",
  "description": "The five security scheme families an A2A Agent Card can declare (API key, HTTP auth, OAuth 2.0, OpenID Connect, mutual TLS), an inference-labelled rule for choosing one, and how a client handles auth-required tasks.",
  "kind": "sub",
  "order": 5,
  "target_query": "a2a authentication",
  "secondary_queries": [],
  "tags": [
    "a2a",
    "authentication",
    "oauth",
    "security-schemes",
    "agent-card",
    "agent2agent"
  ],
  "published": "2026-11-03",
  "updated": "2026-11-03",
  "words": 1123,
  "estimated_tokens": 1494,
  "premium": false,
  "rights": {
    "access": "free",
    "note": "Editorial guides are always free and never part of the licensed corpus.",
    "license": "https://changegamer.ai/license.xml",
    "pricing": "https://changegamer.ai/api/pricing.json",
    "payment": "https://changegamer.ai/api/payment.json"
  },
  "license": "https://changegamer.ai/license.xml",
  "citation": "ChangeGamer (2026-11-03). A2A Authentication Schemes: Which to Declare in Your Agent Card. ChangeGamer. https://changegamer.ai/articles/a2a-authentication-schemes (updated 2026-11-03).",
  "bibtex": "@misc{changegamer_a2a_authentication_schemes, title = {A2A Authentication Schemes: Which to Declare in Your Agent Card}, publisher = {ChangeGamer}, year = {2026}, url = {https://changegamer.ai/articles/a2a-authentication-schemes}, note = {Updated 2026-11-03}}",
  "canonical": "https://changegamer.ai/articles/a2a-authentication-schemes",
  "markdown": "https://changegamer.ai/articles/a2a-authentication-schemes.md",
  "takeaways": [
    "An A2A Agent Card declares how clients must authenticate through securitySchemes plus security requirements, and the A2A reference lists five scheme families: API key, HTTP auth, OAuth 2.0, OpenID Connect and mutual TLS.",
    "A2A v1.0.0 modernised its OAuth 2.0 flows by adding device code and PKCE, according to the A2A reference in the ChangeGamer corpus, which was last updated on 27 October 2026.",
    "The TASK_STATE_AUTH_REQUIRED state pauses an A2A task for the client, so, as inference, a client should treat it as a recoverable request for credentials and not as a failure.",
    "Which authentication scheme production A2A agents actually use could not be confirmed from the corpus, so any claim that one scheme dominates is unverified.",
    "Choosing an A2A scheme is, as inference, a question of who the caller acts for: workload identity suits a service calling in its own right, while delegated authority needs a scoped OAuth grant."
  ],
  "outline": [
    {
      "depth": 2,
      "text": "Which authentication schemes can an A2A Agent Card declare?",
      "anchor": "which-authentication-schemes-can-an-a2a-agent-card-declare",
      "url": "https://changegamer.ai/articles/a2a-authentication-schemes#which-authentication-schemes-can-an-a2a-agent-card-declare"
    },
    {
      "depth": 2,
      "text": "How should you choose a scheme? (inference)",
      "anchor": "how-should-you-choose-a-scheme-inference",
      "url": "https://changegamer.ai/articles/a2a-authentication-schemes#how-should-you-choose-a-scheme-inference"
    },
    {
      "depth": 2,
      "text": "How does a client satisfy a declared scheme?",
      "anchor": "how-does-a-client-satisfy-a-declared-scheme",
      "url": "https://changegamer.ai/articles/a2a-authentication-schemes#how-does-a-client-satisfy-a-declared-scheme"
    },
    {
      "depth": 3,
      "text": "How do securitySchemes relate to the extended card?",
      "anchor": "how-do-securityschemes-relate-to-the-extended-card",
      "url": "https://changegamer.ai/articles/a2a-authentication-schemes#how-do-securityschemes-relate-to-the-extended-card"
    },
    {
      "depth": 2,
      "text": "What should a client do on TASK_STATE_AUTH_REQUIRED?",
      "anchor": "what-should-a-client-do-on-task-state-auth-required",
      "url": "https://changegamer.ai/articles/a2a-authentication-schemes#what-should-a-client-do-on-task-state-auth-required"
    },
    {
      "depth": 2,
      "text": "What mistakes should you avoid when declaring schemes? (inference)",
      "anchor": "what-mistakes-should-you-avoid-when-declaring-schemes-inference",
      "url": "https://changegamer.ai/articles/a2a-authentication-schemes#what-mistakes-should-you-avoid-when-declaring-schemes-inference"
    },
    {
      "depth": 2,
      "text": "What does the corpus not say about A2A authentication?",
      "anchor": "what-does-the-corpus-not-say-about-a2a-authentication",
      "url": "https://changegamer.ai/articles/a2a-authentication-schemes#what-does-the-corpus-not-say-about-a2a-authentication"
    },
    {
      "depth": 2,
      "text": "Does ChangeGamer implement A2A authentication?",
      "anchor": "does-changegamer-implement-a2a-authentication",
      "url": "https://changegamer.ai/articles/a2a-authentication-schemes#does-changegamer-implement-a2a-authentication"
    }
  ],
  "faq": [
    {
      "question": "What authentication does A2A support?",
      "answer": "A2A Agent Cards can declare API key, HTTP auth, OAuth 2.0, OpenID Connect and mutual TLS security schemes, according to the A2A reference in the ChangeGamer corpus. The card lists the schemes and the security requirements a client must meet before calling the agent."
    },
    {
      "question": "Does A2A support OAuth 2.0 device code and PKCE?",
      "answer": "A2A v1.0.0, released on 12 March 2026, modernised its OAuth 2.0 flows by adding device code and PKCE, according to the A2A reference in the ChangeGamer corpus. The reference does not detail the flow mechanics, so read the specification before implementing them."
    },
    {
      "question": "What should an A2A client do when a task becomes auth-required?",
      "answer": "An A2A client should treat TASK_STATE_AUTH_REQUIRED as a pause awaiting credentials, not as a failure. The reference says only that the state pauses the task for the client; how the task resumes is not in the corpus, so that handling pattern is inference."
    },
    {
      "question": "Which authentication scheme do production A2A agents use most?",
      "answer": "Which authentication scheme production A2A agents use most could not be confirmed from the ChangeGamer corpus as of November 2026. The reference lists the supported families but publishes no adoption figures, so treat any popularity claim as unverified."
    }
  ],
  "body": "To pick an A2A authentication scheme, choose from the five families an Agent Card can declare, then make sure a client can satisfy the one you choose. The [A2A protocol reference](/resources/a2a-protocol) (updated 27 October 2026) lists them. The [pillar](/articles/a2a-protocol-in-practice) shows where authentication appears in the protocol; this article is about the decision: which scheme to declare, and what a calling client does to meet it.\n\n## Which authentication schemes can an A2A Agent Card declare?\n\nThe five families (API key, HTTP auth, OAuth 2.0, OpenID Connect and mutual TLS) and the v1.0.0 OAuth changes are listed in the pillar's [authentication section](/articles/a2a-protocol-in-practice), so they are not repeated here. What matters for the choice is what the reference leaves out: it gives no placement or format detail for API keys, no sub-type detail for HTTP auth, no flow detail for OpenID Connect and no certificate detail for mutual TLS. It also does not say how any family maps onto the three bindings; the [bindings article](/articles/a2a-bindings-json-rpc-grpc-rest) covers what differs on the wire, and credential passing per binding is not stated here.\n\n## How should you choose a scheme? (inference)\n\nChoose by who the caller is acting for, because the identity reference splits agent authentication into two layers. Everything in this table is inference, mapping that split onto A2A's families; the specification gives no such guidance.\n\nThe [agent identity reference](/resources/agent-identity-authentication) (updated 17 July 2026) separates workload identity (what the agent is) from delegated authority (permission to act for a human or organisation), and says conflating them causes most agent-auth mistakes.\n\n| Situation | Lean toward (inference) | Reason (inference) |\n|---|---|---|\n| Your own services call a partner agent as themselves | Mutual TLS, or OAuth 2.0 | Workload identity: the caller proves what it is |\n| An agent acts for a human user | OAuth 2.0 or OpenID Connect | Delegated authority needs a scoped grant, not a forwarded user token |\n| A human must approve access interactively on a limited device | OAuth 2.0 device code | v1.0.0 added it for exactly this shape of flow |\n| Early prototype, one known caller | API key | Cheapest to run, but the identity reference warns that static keys shared across instances cannot be pinned to one process or rotated automatically |\n| You already issue enterprise single sign-on | OpenID Connect | Reuses an identity provider you operate |\n\nDeclare only the schemes you will operate and monitor. As inference, a card that lists five schemes forces every caller to guess which one you actually enforce. The identity article covers the underlying distinction, including SPIFFE-issued workload credentials, in [how AI agents prove identity and delegated authority](/articles/agent-identity-and-authentication-for-ai-agents).\n\n## How does a client satisfy a declared scheme?\n\nA client reads the card's `securitySchemes` and security requirements, obtains matching credentials, and presents them before calling. The pillar states the same duty; the corpus gives no further mechanics.\n\nA workable client sequence, all of it inference beyond the first step:\n\n1. Fetch the card from `/.well-known/agent-card.json`.\n2. Read the security requirements and pick a requirement set you can meet.\n3. Obtain credentials out of band or through the declared flow (for OAuth 2.0, the flow the card names).\n4. Cache the credential only for its lifetime; never log it.\n5. Call the agent, and handle an auth-required task as described below.\n\nScheme parameters beyond the family name, such as token endpoints or scopes, are not described in the corpus. Read them from the card and the specification, not from this article.\n\n### How do securitySchemes relate to the extended card?\n\nThe `securitySchemes` block belongs in the public card, because a client must learn how to authenticate before it can (inference). The extended card is the other half: if `extendedAgentCard` is true, an authenticated client can fetch fuller details. The [Agent Card design article](/articles/a2a-agent-card-design) works through that split, so it is not restated here.\n\n## What should a client do on TASK_STATE_AUTH_REQUIRED?\n\nA client should treat `TASK_STATE_AUTH_REQUIRED` as a pause awaiting credentials, not an error. The reference says the state pauses the task for the client; the [task lifecycle article](/articles/a2a-task-lifecycle) places it among the interrupted states.\n\nEverything past \"pauses for the client\" is inference. A reasonable pattern:\n\n- Surface the pause to whoever holds authority. The lifecycle article's rule: supply credentials you are authorised to give, escalate otherwise.\n- Obtain the credential through a flow you control, using the narrowest scope that will do.\n- Resume or retry through the operations the reference lists. The corpus does not describe how a paused task resumes, so test this against the agent you are calling.\n- Bound the number of attempts, so a wrong credential cannot loop.\n\nDo not hand an agent the user's own token to satisfy the pause; the identity reference calls that an anti-pattern because it invites overbroad access and confused-deputy replay.\n\n## What mistakes should you avoid when declaring schemes? (inference)\n\nAvoid declaring a scheme you cannot revoke, rotate or audit; every point here is inference from the identity reference, not A2A specification text.\n\n- **Shared static keys.** One API key handed to every caller cannot be tied to one running process and gives indefinite access if leaked, which is the identity reference's objection to static shared keys. If you start with an API key, issue one per caller.\n- **Credentials in the card.** Neither the public nor the extended card is a place for keys or tokens; the card says how to authenticate, not what to present.\n- **Treating the card as proof of the publisher.** Authentication proves the caller to you. An unsigned card still proves nothing about who published it, and signing is optional.\n- **Over-broad scopes.** For OAuth 2.0 or OpenID Connect, request the narrowest scope that completes the skill being called, and (the identity reference's advice) prefer short token lifetimes with audience binding.\n\n## What does the corpus not say about A2A authentication?\n\nThe corpus does not say which scheme production agents use, how credentials travel per binding, how failures are signalled, or how a task resumes after auth-required. As of 3 November 2026, the A2A reference records no HTTP status mapping, no authentication error codes and no adoption figures, so this article states none.\n\nTwo neighbouring questions are separate. Card signing is optional; authentication proves the caller to the callee, not the reverse. Verifying a remote agent is covered in [trusting a remote agent](/articles/trusting-a-remote-agent), and how credentials propagate across delegation hops in [credential propagation](/articles/credential-propagation-across-agent-hops).\n\n## Does ChangeGamer implement A2A authentication?\n\nNo. As of 2 November 2026, the site speaks MCP over JSON-RPC and implements no A2A endpoint, as the [bindings article](/articles/a2a-bindings-json-rpc-grpc-rest) records, so the guidance above is reasoning from the reference, not from running an A2A security scheme in production.",
  "cluster": {
    "id": "a2a-in-practice",
    "title": "A2A in practice: agent-to-agent interoperability",
    "description": "How to build, secure and run agent-to-agent delegation with the A2A protocol — when A2A beats in-process subagents, Agent Cards, task lifecycle, bindings, authentication, credential propagation across hops, discovery and trust, v0.3 to v1.0 migration, framework support and failure handling — not MCP tool serving (mcp-in-practice) and not single-agent identity (agent-identity-and-authentication-for-ai-agents).",
    "status": "complete",
    "pillar": {
      "slug": "a2a-protocol-in-practice",
      "title": "The A2A Protocol in Practice: Agent Cards, Tasks, Auth and When You Need It",
      "description": "A practical guide to the A2A (Agent2Agent) protocol: what it standardizes, when it beats in-process subagents, how Agent Cards, task states, bindings and authentication work, how credentials cross agent hops, and how to launch it.",
      "kind": "pillar",
      "order": 0,
      "html": "https://changegamer.ai/articles/a2a-protocol-in-practice",
      "markdown": "https://changegamer.ai/articles/a2a-protocol-in-practice.md",
      "json": "https://changegamer.ai/api/articles/a2a-protocol-in-practice.json"
    },
    "articles": [
      {
        "slug": "do-you-need-a2a",
        "title": "Do You Need A2A for a Multi-Agent System? A Decision Rule",
        "description": "A decision rule for choosing between in-process subagents and the A2A protocol: start with one agent, add more only for a named failure, and cross the network with A2A only when a remote party, framework or trust boundary demands it.",
        "kind": "sub",
        "order": 1,
        "html": "https://changegamer.ai/articles/do-you-need-a2a",
        "markdown": "https://changegamer.ai/articles/do-you-need-a2a.md",
        "json": "https://changegamer.ai/api/articles/do-you-need-a2a.json"
      },
      {
        "slug": "a2a-agent-card-design",
        "title": "Designing an Agent Card: What Goes in the Public Card and What Waits",
        "description": "How to design an A2A Agent Card: the field groups, the well-known path, optional signing, and a rule for splitting what the public card advertises from what an authenticated client fetches in the extended card.",
        "kind": "sub",
        "order": 2,
        "html": "https://changegamer.ai/articles/a2a-agent-card-design",
        "markdown": "https://changegamer.ai/articles/a2a-agent-card-design.md",
        "json": "https://changegamer.ai/api/articles/a2a-agent-card-design.json"
      },
      {
        "slug": "a2a-task-lifecycle",
        "title": "A2A Task States and Operations: What Each State Means for an Orchestrator",
        "description": "The eight A2A task states, the eleven operations that act on tasks, the difference between interrupted and terminal states, and an inference-labelled handler matrix for an orchestrator.",
        "kind": "sub",
        "order": 3,
        "html": "https://changegamer.ai/articles/a2a-task-lifecycle",
        "markdown": "https://changegamer.ai/articles/a2a-task-lifecycle.md",
        "json": "https://changegamer.ai/api/articles/a2a-task-lifecycle.json"
      },
      {
        "slug": "a2a-bindings-json-rpc-grpc-rest",
        "title": "A2A JSON-RPC vs gRPC vs REST: What Differs Between the Three Bindings",
        "description": "The three A2A protocol bindings (JSON-RPC, gRPC, HTTP+JSON/REST): what the specification keeps identical, what differs on the wire, and an inference-labelled guide to picking one.",
        "kind": "sub",
        "order": 4,
        "html": "https://changegamer.ai/articles/a2a-bindings-json-rpc-grpc-rest",
        "markdown": "https://changegamer.ai/articles/a2a-bindings-json-rpc-grpc-rest.md",
        "json": "https://changegamer.ai/api/articles/a2a-bindings-json-rpc-grpc-rest.json"
      },
      {
        "slug": "a2a-authentication-schemes",
        "title": "A2A Authentication Schemes: Which to Declare in Your Agent Card",
        "description": "The five security scheme families an A2A Agent Card can declare (API key, HTTP auth, OAuth 2.0, OpenID Connect, mutual TLS), an inference-labelled rule for choosing one, and how a client handles auth-required tasks.",
        "kind": "sub",
        "order": 5,
        "html": "https://changegamer.ai/articles/a2a-authentication-schemes",
        "markdown": "https://changegamer.ai/articles/a2a-authentication-schemes.md",
        "json": "https://changegamer.ai/api/articles/a2a-authentication-schemes.json"
      },
      {
        "slug": "credential-propagation-across-agent-hops",
        "title": "Credential Propagation Across Agent Hops: Exchange, Audience and Depth",
        "description": "How to carry authority through a chain of delegating agents: why forwarding a token fails, per-hop RFC 8693 exchange with act and may_act, RFC 8707 audience binding, a worked sequence and a failure walk-through.",
        "kind": "sub",
        "order": 6,
        "html": "https://changegamer.ai/articles/credential-propagation-across-agent-hops",
        "markdown": "https://changegamer.ai/articles/credential-propagation-across-agent-hops.md",
        "json": "https://changegamer.ai/api/articles/credential-propagation-across-agent-hops.json"
      },
      {
        "slug": "trusting-a-remote-agent",
        "title": "How to Verify a Remote A2A Agent Before Delegating",
        "description": "What a signed Agent Card proves and does not, how domain, key and ERC-8004 identity differ, where Web Bot Auth fits, and why card text and returned artifacts are untrusted input, ending in an ordered checklist.",
        "kind": "sub",
        "order": 7,
        "html": "https://changegamer.ai/articles/trusting-a-remote-agent",
        "markdown": "https://changegamer.ai/articles/trusting-a-remote-agent.md",
        "json": "https://changegamer.ai/api/articles/trusting-a-remote-agent.json"
      },
      {
        "slug": "a2a-agent-discovery-without-a-registry",
        "title": "How Do Agents Discover A2A Agents Without a Standard Registry?",
        "description": "The three discovery strategies in the A2A documentation, why the spec prescribes no registry API, what vendor registries fill the gap, and how to choose between a well-known URI, a registry and direct configuration.",
        "kind": "sub",
        "order": 8,
        "html": "https://changegamer.ai/articles/a2a-agent-discovery-without-a-registry",
        "markdown": "https://changegamer.ai/articles/a2a-agent-discovery-without-a-registry.md",
        "json": "https://changegamer.ai/api/articles/a2a-agent-discovery-without-a-registry.json"
      },
      {
        "slug": "a2a-v0-3-to-v1-0-migration",
        "title": "Migrate A2A v0.3 to v1.0: An Ordered Checklist for Clients and Servers",
        "description": "An ordered checklist for moving an A2A client or server from v0.3 to v1.0: the A2A-Version header rule, method names, Part shape, a runnable grep check, and symptom-to-cause pointers.",
        "kind": "sub",
        "order": 9,
        "html": "https://changegamer.ai/articles/a2a-v0-3-to-v1-0-migration",
        "markdown": "https://changegamer.ai/articles/a2a-v0-3-to-v1-0-migration.md",
        "json": "https://changegamer.ai/api/articles/a2a-v0-3-to-v1-0-migration.json"
      },
      {
        "slug": "agent-frameworks-with-a2a-support",
        "title": "Does Your Agent Framework Really Support A2A? How to Read the Claim and What to Test",
        "description": "Which agent frameworks list A2A support, what native, add-on, platform-gated and unrecorded mean, a short test checklist to run before you rely on it, and how that differs from the protocol's own SDKs.",
        "kind": "sub",
        "order": 10,
        "html": "https://changegamer.ai/articles/agent-frameworks-with-a2a-support",
        "markdown": "https://changegamer.ai/articles/agent-frameworks-with-a2a-support.md",
        "json": "https://changegamer.ai/api/articles/agent-frameworks-with-a2a-support.json"
      },
      {
        "slug": "failure-handling-across-agent-boundaries",
        "title": "Failed or Rejected Remote A2A Task: Retry, Degrade or Surface?",
        "description": "How to decide what happens after a remote A2A task ends badly: the side-effects test that splits failed into retry or surface, why rejected is not a retry case, why input-required is not a failure, and a structured result for the caller.",
        "kind": "sub",
        "order": 11,
        "html": "https://changegamer.ai/articles/failure-handling-across-agent-boundaries",
        "markdown": "https://changegamer.ai/articles/failure-handling-across-agent-boundaries.md",
        "json": "https://changegamer.ai/api/articles/failure-handling-across-agent-boundaries.json"
      },
      {
        "slug": "a2a-production-launch-checklist",
        "title": "A2A Production Launch Checklist: Gates to Pass Before You Ship",
        "description": "A go/no-go checklist for launching A2A agent-to-agent delegation, one line per gate, each pointing to the article that owns the detail: need, card, states, binding, auth, trust, versioning, failure.",
        "kind": "sub",
        "order": 12,
        "html": "https://changegamer.ai/articles/a2a-production-launch-checklist",
        "markdown": "https://changegamer.ai/articles/a2a-production-launch-checklist.md",
        "json": "https://changegamer.ai/api/articles/a2a-production-launch-checklist.json"
      }
    ]
  },
  "navigation": {
    "pillar": {
      "slug": "a2a-protocol-in-practice",
      "title": "The A2A Protocol in Practice: Agent Cards, Tasks, Auth and When You Need It",
      "description": "A practical guide to the A2A (Agent2Agent) protocol: what it standardizes, when it beats in-process subagents, how Agent Cards, task states, bindings and authentication work, how credentials cross agent hops, and how to launch it.",
      "kind": "pillar",
      "order": 0,
      "html": "https://changegamer.ai/articles/a2a-protocol-in-practice",
      "markdown": "https://changegamer.ai/articles/a2a-protocol-in-practice.md",
      "json": "https://changegamer.ai/api/articles/a2a-protocol-in-practice.json"
    },
    "previous": {
      "slug": "a2a-bindings-json-rpc-grpc-rest",
      "title": "A2A JSON-RPC vs gRPC vs REST: What Differs Between the Three Bindings",
      "description": "The three A2A protocol bindings (JSON-RPC, gRPC, HTTP+JSON/REST): what the specification keeps identical, what differs on the wire, and an inference-labelled guide to picking one.",
      "kind": "sub",
      "order": 4,
      "html": "https://changegamer.ai/articles/a2a-bindings-json-rpc-grpc-rest",
      "markdown": "https://changegamer.ai/articles/a2a-bindings-json-rpc-grpc-rest.md",
      "json": "https://changegamer.ai/api/articles/a2a-bindings-json-rpc-grpc-rest.json"
    },
    "next": {
      "slug": "credential-propagation-across-agent-hops",
      "title": "Credential Propagation Across Agent Hops: Exchange, Audience and Depth",
      "description": "How to carry authority through a chain of delegating agents: why forwarding a token fails, per-hop RFC 8693 exchange with act and may_act, RFC 8707 audience binding, a worked sequence and a failure walk-through.",
      "kind": "sub",
      "order": 6,
      "html": "https://changegamer.ai/articles/credential-propagation-across-agent-hops",
      "markdown": "https://changegamer.ai/articles/credential-propagation-across-agent-hops.md",
      "json": "https://changegamer.ai/api/articles/credential-propagation-across-agent-hops.json"
    }
  },
  "resources": [
    {
      "slug": "a2a-protocol",
      "html": "https://changegamer.ai/resources/a2a-protocol",
      "markdown": "https://changegamer.ai/resources/a2a-protocol.md",
      "json": "https://changegamer.ai/api/resources/a2a-protocol.json"
    },
    {
      "slug": "agent-identity-authentication",
      "html": "https://changegamer.ai/resources/agent-identity-authentication",
      "markdown": "https://changegamer.ai/resources/agent-identity-authentication.md",
      "json": "https://changegamer.ai/api/resources/agent-identity-authentication.json"
    }
  ]
}