Agent Runtimes and kagent

How an agent runtime starts, stops, and isolates work on Kubernetes, including kagent and Agent Substrate.

An agent runtime is the thing that starts, stops, and isolates an agent while it works. kagent puts that on Kubernetes.

These posts cover the harness, the substrate that packs many agents onto few nodes, and what breaks when isolation is shared. Identity for those agents is in Agent Identity .

Start here

  1. What Is a Cloud-Native Agent Harness? (kagent) When a desktop coding harness stops fitting: what a cloud-native agent harness is, and how kagent splits harness, template, instance and sandbox on Kubernetes.
  2. kagent Agent Substrate: Thousands of Agents kagent Agent Substrate is built to run thousands of AI agents on tens of pods. I stood it up on kind: snapshot restores, autoscaling, and every gotcha I hit.
  3. Agent Runtimes Have an Autoscaler, Not a Scheduler Agent runtimes optimize for density and reject when full. That is not scheduling. Three failure modes at capacity, and who gets to call an agent critical.

Every post in this hub

  1. How AI Agents Sharing a Sandbox Leak Each Other's Data
    A finance AI agent leaked bank balances into company Slack by sharing compute with another agent. Reproduced in Docker, then fixed with separate sandboxes.
  2. What Is a Cloud-Native Agent Harness? (kagent)
    When a desktop coding harness stops fitting: what a cloud-native agent harness is, and how kagent splits harness, template, instance and sandbox on Kubernetes.
  3. Agent Runtimes Have an Autoscaler, Not a Scheduler
    Agent runtimes optimize for density and reject when full. That is not scheduling. Three failure modes at capacity, and who gets to call an agent critical.
  4. Regression Tests for kagent Agents with agentevals
    kagent records agent sessions as ADK events. agentevals scores chat-message trajectories. I built the bridge, and a suite that catches an ungrounded agent.
  5. kagent v1.0: Claude Code Harness on Kubernetes
    kagent v1.0.0-alpha1 swaps its agent API for AgentInstance: checkpoint and fork a conversation, resume from a golden snapshot, run Claude Code as the harness.
  6. Substrate Scope: a Live Visualizer for Agent Substrate
    Substrate Scope is an open source live visualizer for Agent Substrate: worker bays, restore queues, snapshot storage, telemetry, and per-agent activity.
  7. Revenant: A Gateway-Native, Always-On Agent Runtime
    Revenant is an always-on agent runtime in Rust where an AI gateway owns every key, budget, and failover, and a hard liveness rule means no turn goes silent.
  8. kagent Agent Substrate: Thousands of Agents
    kagent Agent Substrate is built to run thousands of AI agents on tens of pods. I stood it up on kind: snapshot restores, autoscaling, and every gotcha I hit.

Frequently asked questions

What is an agent runtime?

An agent runtime is the thing that starts, stops, and isolates an agent while it works. On Kubernetes that includes the harness, the sandbox, and the pool of workers that actually run the loop.

What is kagent Agent Substrate?

Agent Substrate is kagent’s way of packing many agents onto fewer nodes. It snapshots a warm worker, parks idle agents, and brings them back without paying the full cold-start cost each time.