a2a cloud journal

Field notes for autonomous software.

Architecture notes, launch updates, and hard-won details from building runtime infrastructure for agents that ship, earn, and prove their work.

latest

Oct 1, 2026

Neon vs an auto-provisioned agent database: which Postgres should your AI agent use?

Neon is excellent Postgres. But for a LangGraph, CrewAI, or OpenAI Agents SDK agent, the real choice is between a database you provision and wire yourself and one that ships with the agent: scoped credential, lifecycle-coupled, with MCP, an API, and signed receipts alongside it. An honest comparison, including when Neon wins.

read note →

All posts

a2acloud.io ↗

Oct 1, 2026·7 min read·latest

Neon vs an auto-provisioned agent database: which Postgres should your AI agent use?

postgresneoncomparisonlanggraphcrewaiopenai-agents-sdkmcpsigned-receiptsagent-infrastructure

Sep 30, 2026·5 min read

Cloudflare Agents Alternative for Python Agents: Postgres, MCP, and Signed Receipts Included

Cloudflare's Agents SDK is great for TypeScript agents on Workers. If you built on LangGraph, CrewAI, or the OpenAI Agents SDK in Python and need a shared Postgres database, MCP tools, and verifiable proof of what your agent did, here's an honest comparison.

cloudflare agents alternativecomparisonlanggraphcrewaiopenai agents sdkpostgresmcpsigned receipts

Sep 29, 2026·5 min read

Add memory and RAG to an AI agent with Postgres and pgvector

Agent memory is three different things — thread state, long-term facts, and retrieval. You can run all three in one Postgres database. Here's how, with LangGraph checkpoints and pgvector, on a database your agent gets at deploy time.

agent memoryragpgvectorpostgreslanggraphtutorial

Sep 25, 2026·5 min read

Expose Your AI Agent as an API (With an OpenAPI Spec You Didn't Write)

Turning an agent into an API usually means a FastAPI wrapper, duplicated request models, and a hand-maintained OpenAPI file that drifts. On a2a cloud every typed skill becomes POST /invoke/{skill} with a generated OpenAPI 3.1 contract, a typed TS client, MCP, and a signed receipt per call.

ai-agentsapiopenapimcplanggraphcrewaiopenai-agents-sdkcapability

Aug 24, 2026·3 min read

Give your AI agent a Postgres database

Every agent deployed to a2a cloud is provisioned with its own managed Postgres — injected as DATABASE_URL, isolated per agent, and covered by signed receipts.

postgresagent memoryai agentsdatabaselanggraphcrewaiopenai agents sdkcapabilitya2a cloudsigned receipts

Aug 22, 2026·4 min read

LangSmith vs a2a cloud: tracing is not an audit trail

Tracing answers why your agent did something. Signed receipts prove what it did and who authorized it. Where LangSmith and a2a cloud actually differ — and why most teams want both.

agent audit trailsigned receiptslangsmithobservabilitycomparison

Aug 20, 2026·7 min read

SOC 2 Evidence for AI Agents: What Auditors Actually Ask For

SOC 2 isn't a policy exercise — it's an evidence hunt. When the thing under audit is an autonomous agent, most of your evidence is scattered application logs you're asking the auditor to trust. Here's what evidence an agent actually needs, and why a signed receipt beats a log export.

ai agentssoc 2complianceagent audit trailsigned receiptsgovernanceagent securitya2a cloud

Aug 19, 2026·4 min read

Wire an MCP Server to Your Agent (With Zero Config)

The Model Context Protocol lets any client call your agent as a tool — but only if something hosts, authenticates, and keeps the MCP server in sync with your code. On a2a cloud that server is generated from your agent's skills on every deploy. Here's how it works.

mcpmodel context protocolai agentsagent toolstool callingcapabilitya2a cloudagent infrastructure

Jul 19, 2026·7 min read

From Prompt to Proof: Rebuilding Agent Studio Around Working Software

How a2a cloud changed Agent Studio from a code generator into a production system that proves frontends, backends, artifacts, browser workflows, funded trials, and public distribution actually work.

Jul 12, 2026·5 min read

Deploy an OpenAI Agents SDK Agent in 60 Seconds (With Postgres, MCP, and Signed Receipts)

You built an agent with the OpenAI Agents SDK in an afternoon. Getting it into production — persistent state, tools, an API, and proof of what it did — is the other two weeks. On a2a cloud that infrastructure is provisioned with the agent on first deploy.

openai agents sdkdeploy ai agentai agentspostgresmcpsigned receiptsquickstarta2a cloud

Jul 11, 2026·3 min read

Your agent has an email address now

Every a2a cloud agent can now have a real inbox: one line of YAML, threaded replies in seconds, default-deny senders, send caps, and an audit event for every mail.

emailagent inboxmailboxesproductdeveloper experience

Jul 8, 2026·3 min read

Every agent is now a CLI: a2a use

`a2a use <agent>` turns any published agent into typed terminal commands — real flags from the card's JSON Schema, @file and stdin support, and a setup handshake for keys and config.

a2a useagent to cliclideveloper experience

Jul 7, 2026·3 min read

a2a dev now runs your agent in the cloud

`a2a dev` keeps the local inner loop but moves the runtime to the cloud: edit locally, run your agent on its scale-to-zero dev box, hot-reload on save, and share it on a public URL.

a2a devcloud devdeveloper experiencecli

Jul 5, 2026·5 min read

Compliance is a query, not a quarter

EU AI Act enforcement lands 2 August 2026. Today we ship Article 12 decision records, floored retention, risk classification, and one-click evidence packs - as a view over receipts your agents already sign, not a second store.

eu ai actcompliancesigned receiptsgovernanceai agentsproduct

Jul 5, 2026·3 min read

Introducing A2A Cloud Academy: four free courses from a2a init to production

A2A Cloud Academy is live: four free, in-depth courses — your first deployed agent, the skill runtime, production operations, and frontends/OpenAPI/MCP — 53 lessons grounded in the shipping platform.

academytutorialsplatformagents

Jul 5, 2026·4 min read

Deploy a CrewAI Agent in 60 Seconds (With Postgres, MCP, and Signed Receipts)

You built a CrewAI crew. Now it needs a database, tools, an API, and proof it ran safely. On a2a cloud all four are provisioned with the agent on the first deploy. Here's the whole path.

crewaideploy ai agentai agentspostgresmcpsigned receiptsquickstarta2a cloud

Jul 2, 2026·6 min read

Signed Receipts vs Trace Logs: The Agent Audit Trail That Actually Proves What Happened

A trace log is something you're asked to trust. A signed receipt is something you can verify. For AI agents acting on their own, that difference decides whether your audit trail survives a real dispute — or an auditor.

ai agentssigned receiptsagent audit trailtrace logseu ai actsoc 2governancea2a cloud

Jul 1, 2026·3 min read

SSH into any agent, in one command

a2a ssh <agent> drops you into a throwaway cloud box with Node, Python, a2a-pack, and your agent's repo already loaded — over your existing ingress, VS Code-ready, and scaled to zero when idle.

productdeveloper-experienceagents

Jun 29, 2026·3 min read

Our Growth Engine Is an Agent on a2a Cloud — Here's the Signed Receipt

We don't run our go-to-market with a stack of SaaS tools. We run it with an agent deployed on a2a cloud — the same platform we sell. Every move it makes leaves a signed receipt. Here's one.

dogfoodingsigned receiptsai agentsgrowtha2a cloudproof

Jun 29, 2026·4 min read

Deploy a LangGraph Agent in 60 Seconds (With Postgres, MCP, and Signed Receipts)

You wrote a LangGraph graph. Now it needs a database, tools, an API, and proof it ran safely. On a2a cloud all four are provisioned with the agent on the first deploy. Here's the whole path.

langgraphdeploy ai agentai agentspostgresmcpsigned receiptsquickstarta2a cloud

Jun 27, 2026·3 min read

How to Give AI Agents Database Access Without Handing Over Production

Most stacks only offer agents one DB access pattern: a single connection string with full access to prod. Here's the safer model — scoped per-agent grants, human approval on risky steps, and a signed receipt for every action.

ai agentsagent securitydatabasemcpproduction

Jun 27, 2026·5 min read

How a2a cloud isolates agents

A high-level look at how a2a cloud combines signed grants, bounded workspaces, microVM sandboxes, and scoped source access so agents can act without inheriting unlimited authority.

agentssecurityinfrastructure

Jun 20, 2026·8 min read

Early Investigation: Dynamic Memory for Agents

A technical look at a schema-aware semantic memory system where Pydantic models, tool calls, agent traces, vector search, graph edges, and feedback loops become trainable memory.

agentsmemorysemantic grapha2amcpvector databasegraph databasemachine learningpydantic

Jun 15, 2026·6 min read

The Cloud Wasn't Built for Autonomous AI Agents

Modern cloud platforms were designed for deterministic microservices. Autonomous AI agents need stronger isolation, agent-native networking, and auditable decision trails.

a2a cloudai agentscloud infrastructuremcpa2asecuritygovernance

Jun 14, 2026·4 min read

How A2A Generates Agents from OpenAPI Specs

A look inside the A2A OpenAPI generator: how it reads a spec, turns routes into tools, wires auth and egress, and deploys an editable agent.

openapiagentsa2aengineering

Jun 12, 2026·4 min read

Why Every Agent Gets Its Own Repo

Every agent is software. Giving each agent its own repo makes it easier to understand, edit, review, deploy, and improve without stepping on unrelated work.

a2aagentsdeveloper-experiencegitplatform

Jun 11, 2026·6 min read

A2A for Developers: Deploy an Agent in 15 Minutes with a2a cloud

Use the A2A Pack template to scaffold a typed agent, run it locally, generate OpenAPI and TypeScript clients, and deploy it to a2a cloud.

a2aa2a-packagentstutorialopenapidevelopers

Jun 11, 2026·3 min read

Making Argo CD and Knative Services Play Nicely

Knative Services are controller-owned resources, so Argo CD needs health checks, diff ignores, and careful sync settings to avoid fighting the platform.

kubernetesargocdknativeplatformoperations

Jun 6, 2026·5 min read

Databases for Production AI Agents

a2a cloud is making Postgres a first-class agent resource: declared by agents, governed by the platform, durable at rest, and able to suspend compute when idle.

agentsdatabasespostgreskubernetesplatform

Jun 3, 2026·5 min read

The Kernel: Where Agent Simulations Become Product Infrastructure

A2A Cloud's Kernel turns agents from isolated tools into participants in bounded, replayable simulations: markets, tournaments, cohorts, arenas, resource competitions, and future agent economies.

a2a cloudkernelagent simulationsmulti-agent systemsagentsplatform engineeringagent economies

Jun 2, 2026·4 min read

Meta-Agents, Part 2: The Kernel Behind the Team

If Part 1 was the big idea, Part 2 is the backstage tour. A meta-agent is not one giant chatbot. It is a smart coordinator that keeps a goal in view, brings in the right specialist agents, remembers what matters, and keeps moving until the job is done.

a2a cloudmeta-agentsagentsmemorycompositionplatform engineering

Jun 2, 2026·6 min read

Agent Studio: Why Agents Need a Build, Review, and Improve Loop

Building an agent is only step one. Agent Studio is A2A Cloud's loop for creating an agent, checking whether it works, improving it, and handing back a clear report instead of a mystery result.

a2a cloudagent studioagent qualityagentsautomation

May 28, 2026·5 min read

Agent-Builder: The Meta-Agent That Writes, Tests, and Ships Other Agents

Describe an agent. Get a live URL. Agent-Builder is the meta-agent inside A2A Cloud that scaffolds, writes, sandboxes, deploys, and verifies new agents — powered by DeepAgents and the platform's own runtime.

a2a cloudagent builderdeepagentsmeta-agentcode generationagentsplatform engineering

May 28, 2026·5 min read

Streaming UX for Long-Running Agents: SSE, Approvals, and Human-in-the-Loop

Real agents do work that takes minutes. They ask questions. They request more scope. They write files mid-task. A2A Cloud's streaming protocol turns every one of those into a live UI event — so users stay in the loop, not in the dark.

a2a cloudstreamingsseuxhuman-in-the-loopagentsfrontend

May 28, 2026·5 min read

Scoped Grants: Agent-to-Agent Delegation Without Over-Privileging

Multi-agent systems break when one agent has to trust another with everything. A2A Cloud ships Ed25519-signed grants that scope workspace access, budget, and runtime per handoff — so trust is bounded, every time.

a2a cloudsecuritygrantsed25519multi-agentdelegationagentsplatform engineering

May 28, 2026·5 min read

Build Your First A2A Agent: From Python Class to Live URL

A Python class, a YAML file, one CLI command. That's the path from "I have an idea" to a live, callable agent on A2A Cloud. Here's exactly how it works.

a2a clouda2a packagentssdkdeveloper experiencegetting startedpython

May 28, 2026·5 min read

Sandbox Isolation: How A2A Cloud Runs Agent Code Inside microVMs

Containers aren't enough. A2A Cloud runs every agent workload inside libkrun microVMs with FUSE-backed workspaces, signed grants, and locked-down egress — so agents can do real work without ever touching the host.

a2a cloudsecuritylibkrunmicrosandboxfuseagentsplatform engineering

May 28, 2026·4 min read

Scaling AI Agents With Argo: From Prototype to Production Fleet — Without the Pain

How a2a cloud uses Argo-driven deployments to turn custom agents into reliable, observable, production-ready services teams can scale with absolute confidence.

agentsargokubernetesagent marketplacea2a cloud

May 28, 2026·5 min read

A2A Agents Just Got a Bigger Platform — Elastic, Async, GitOps-Native

A2A Cloud now ships the foundation for elastic HTTP agents, async KEDA workers, GitOps rollouts, scale-to-zero, and fleet-level growth. The platform shape just changed.

a2a cloudknativekedaagentsplatform engineeringkubernetesscale-to-zero

May 28, 2026·7 min read

LiteLLM + Microsandbox: A2A Cloud Turns Ideas Into Deployed Agents — Fast

Describe an agent. Get a live URL. A2A Cloud fuses LiteLLM routing, scoped workspaces, microsandbox execution, and GitOps deploys into one ridiculously fast build-to-ship loop.

a2a cloudlitellmmicrosandboxagent builderagentsplatform engineeringgitops