NOTES / PART 2 · BUILDING AGENTS

06Harnesses

Everything around the model that turns it into a working agent.

THIS SECTION COVERS

Agent = model + harness, the execution loop, tool registry and tool design, MCP and A2A, state and checkpoints, hooks, subagents, case studies such as coding agents.

Related: Frameworks and SDKs · Safety and guardrails

TOPICS · 9/9 WRITTEN

  1. 06.1Agent = model + harnessWhat a harness is, what goes in it, and why most of the engineering in an agent sits outside the model.
  2. 06.2The execution loopThe loop at the centre of every harness: call the model, run the tools it asks for, feed the results back, and stop when it asks for nothing more.
  3. 06.3Tool design and the tool registryTools are a contract between deterministic software and a non-deterministic caller. How to design them for agents, and how to improve them with evals.
  4. 06.4Model Context Protocol (MCP)The open protocol that lets an agent connect to tools and data through a standard interface: hosts, clients, servers and what servers can offer.
  5. 06.5Agent-to-agent protocols (A2A)How agents built by different teams on different frameworks discover each other and hand work back and forth, and how that differs from MCP.
  6. 06.6State and checkpointsHow a harness saves progress so a run can pause, resume or recover: checkpoints within a run, files across context windows, and what to persist.
  7. 06.7Hooks and lifecycle eventsDeterministic code that runs at fixed points in an agent's lifecycle: how hooks differ from prompts, what events exist, and what to use them for.
  8. 06.8Subagents and delegationHanding parts of a task to helper agents that work in their own context window and report back a summary: when it helps, what it costs, and how to brief them.
  9. 06.9Case study: coding agentsHow the harness ideas come together in a coding agent: the gather, act, verify loop, the tools it needs, why tests make good feedback, and where such agents struggle.