a2a cloud
architecture

Five planes. One agent surface.

a2a cloud is a closed loop: callers hit a protocol edge, the runtime executes against scoped grants, the control plane mints + verifies + records, durable stores keep the proof. Same primitives across every agent.

a2a cloud · platform architecture
live
untrusted
callers
browser / SPA
user session
MCP client
Claude · Cursor · …
A2A peer
agent → agent
HTTP API
curl · script · webhook
request
edge
protocol surface
/app
packed frontend
/mcp
Model Context Protocol
/tasks · /sse
A2A · streaming
AgentCard · /invoke
discovery · REST
route
managed
agent runtime
service identity
per-agent svc account
grant verifier
HMAC · audience · ttl
sandbox
fs / net jail · seccomp
workspace client
scoped files only
skill dispatch
@skill → endpoint + MCP tool
SSE bus
progress · artifacts · q&a
verify · mint · log
platform
control plane
auth
OAuth/OIDC · sessions
registry
agents · cards · versions
deploys
package · build · release
grants
mint · sign · revoke
approvals
scope expand · review
billing
usage ledger · per-call
persist
durable
storage + ops
workspace store
object · versioned
receipts ledger
content-addressed
OCI registry
agent images
audit log
tamper-evident
metrics + traces
OpenTelemetry
closed loop

Every call returns a receipt anchored to a control-plane record.

single identity

The same service identity carries through callers, runtime, and ledger.

durable trail

Receipts + audit log persist independent of the agent image.

5 planes·26 subsystems·1 protocol surfaceread top → bottom · planes own each other below them
deploy flow

Vercel-shaped DX. Production-shaped ops.

Developer gets one command. Platform team gets a reproducible release path, service identity, runtime boundary, protocol surface, health trail.

  1. 01a2a deploy packages source and metadata.
  2. 02Control plane builds and releases the agent service.
  3. 03Packed frontend (if present) is built and mounted at /app.
  4. 04Identity, permissions, docs, protocol endpoints generated.
  5. 05Managed runtime exposes the agent through A2A, MCP, HTTPS.
  6. 06Receipts, evals, artifacts, distribution metadata accrue.
control plane

Six subsystems. One backend.

The control plane is what makes the runtime trustworthy. Each subsystem owns a narrow slice and emits a structured event trail. No subsystem talks to agent code directly — they talk to the runtime.

auth
owns · users · orgs · sessions · tokens
emits · audit · token-rotation
registry
owns · agents · cards · versions · visibility
emits · AgentCard hash · discovery
deploys
owns · build · OCI image · release
emits · release events · image SBOM
grants
owns · mint · sign · revoke · rotation
emits · grant manifests · audit
approvals
owns · scope expansion · human review
emits · approval events · reviewer id
billing
owns · usage ledger · pricing · invoices
emits · per-call rows · payouts
runtime internals

What's inside the agent process.

Each deployed agent runs a small set of runtime components wrapping user code. The shape is the same for every agent — so security guarantees don't depend on what the developer remembered to wire up.

service identity

Per-agent service account. Signs all platform RPCs. Carries through to receipts.

grant verifier

HMAC + audience + expiry + path glob. Runs on every workspace op, not just task start.

sandbox

Filesystem jail, network egress allowlist, seccomp profile, no host secrets in env.

workspace client

Only path agents reach files. Glob-filtered. Writes pinned to output prefix.

skill dispatch

@skill decorator → REST endpoint + MCP tool + AgentCard entry. One source of truth.

SSE bus

Streams progress · artifacts · question-required · status. Buffered for replay.

user code lives above the runtime · grants + sandbox live below it · workspace + control plane live beyond it
protocol surface

Every URL the runtime exposes.

One agent. Many speakers. Each endpoint is the same logical surface in a different protocol. The runtime owns routing, auth, and grant verification at every entry point.

pathspeakerreturnsauth
/.well-known/agent-card.jsondiscovery · registries · MCP gatewaysAgentCard JSONpublic
/tasksA2A peer · planner agenttask id · 202 · SSE linkbearer · service
/tasks/{id}A2A peertask state · artifactsbearer · service
/sse/{task}A2A peer · /appstream: events · artifactsbearer · service
/mcpClaude Code · Cursor · MCP clientsMCP server (tools/list, call)bearer · user
/invoke/{skill}curl · webhook · scriptsskill result JSONbearer · API key
/appbrowserpacked React appsession · inherit
/app/config.jsonbrowser · /appendpoints · schemassession
/auth/sessionbrowseruser · org · scopessession
/runs/{id}/receiptdashboard · auditorreceipt JSONowner · approver
observability

What every run emits.

You don't add observability — the runtime emits it. Receipts, audit events, SSE streams, OTel traces, billing rows: each triggered by a known state transition, each consumed by a known surface.

triggeremitsconsumed by
task acceptedaudit event · run_id · grant_idaudit log · dashboard
file readworkspace op · path · bytes · hashreceipt · replay
file writtenworkspace op · path · bytes · prefix-checkreceipt · replay
skill progressSSE event · stage · % · message/app · A2A peer · MCP
scope requestapproval event · requested paths · reasonapprover inbox · audit
scope grantedrotated grant · prior_id · new_idreceipt · audit
task completeresult · artifact paths · elapsed · statusreceipt · billing · eval
eval scoredrubric · score · failuresreceipt · registry stats
review signedapprover id · ts · notesreceipt · compliance pack
receipt sealedcontent hash · signer · GL anchorledger · auditor · ledger row
open standards

Built on protocols, not lock-in.

Every callable surface is an open standard. Agents that work here work on other A2A and MCP runtimes — and other A2A and MCP services work here.

A2A (Agent2Agent)
aligned
tasks · messages · artifacts · cards · streaming

Open protocol for agent ↔ agent calls. We implement the full surface.

MCP (Model Context Protocol)
aligned
tools · resources · prompts

Every skill is also an MCP tool. Use any deployed agent from any MCP client.

OCI
aligned
agent images · registry

Agents ship as standard OCI images. Compatible with any conformant registry.

OpenAPI / JSON Schema
aligned
skill IO · AgentCard schemas

Skill signatures become JSON Schema; clients can codegen.

OAuth 2.1 / OIDC
aligned
user auth · service identity

Standard token model. No bespoke session formats.

OpenTelemetry
aligned
traces · metrics

Runtime emits OTel; pipe into Datadog, Honeycomb, Grafana, or self-host.

mental model

If you know these, you know this.

a2a cloud borrows shape from infra you already trust — and points it at agents instead of containers, payments, or web workers.

Vercel
like · git push → live URL
here · but for agents, with grants + receipts
Stripe
like · primitives that compose · per-call ledger
here · but the actor is software, not a card
Cloudflare
like · everything behind one runtime edge
here · but the workload is an agent, not a worker
Kubernetes
like · scheduling + identity + secrets
here · but agent-shaped, not container-shaped
deeper

Want to see the rest of the surface?

Architecture pages describe the runtime. Workflows pages describe what teams ship on it. Security explains the trust model. Pick the lens that matches your question.