Source: AI Engineering with Claude (Udacity ND7426), plus current Claude docs I fetched: Claude Code hooks, hooks guide, Agent SDK hooks, permissions, subagents, custom tools, agent loop, Claude Code sub-agents, dynamic workflows, agent teams, and permission modes

Part 5 ended with a debt: autonomy I am willing to grant, and guardrails that hold even when the model would rather keep going.

This is Part 6 of 6 in my Agentic Coding with Claude Code series. The working model I am keeping is simple: eval tells me whether the loop did the job. Guardrails decide what the loop is allowed to do even when the model wants another hop. Bounded autonomy is the grant: which agents, which tools, how many turns, when a human is required.

This is the series closer. I will not rewrite Parts 1 through 5. I will bound the loop those posts already built.

Read More

Source: AI Engineering with Claude (Udacity ND7426), plus current Claude docs I fetched: Agent SDK structured outputs, agent loop, Python ResultMessage, cost tracking, Agent SDK observability, Claude Code monitoring, platform structured outputs, stop reasons, and tool use

Part 4 ended with a debt: whether the loop is actually doing the job. When a skip was correct, when a tool lied, and when the audit trail is the only honest record.

This is Part 5 of 6 in my Agentic Coding with Claude Code series. The working model I am keeping is simple: evaluation is how I score a run. Observability is how I see the run. Structured output is the seam between them.

A harness that cannot tell success from a polite hallucination is a demo. Later I will talk about bounded autonomy. This one stays on scoring the loop and seeing it.

Read More

Source: AI Engineering with Claude (Udacity ND7426), plus current MCP / Claude docs I fetched: architecture, server concepts, client concepts, transports, 2026-07-28 spec notes, Claude Code MCP, Agent SDK MCP, custom tools, and permissions

Part 3 ended with a debt: tools that live outside the process, MCP servers, how they enter the window, and how to keep that surface small.

This is Part 4 of 6 in my Agentic Coding with Claude Code series. The working model I am keeping is simple: the model still reasons. The harness still decides. MCP is how tools and context live outside the process and still enter the window as a governed surface.

Later posts will cover evaluation and observability, then bounded autonomy. This one stays on the MCP boundary.

Read More

Source: AI Engineering with Claude (Udacity ND7426), plus current Claude docs I fetched: Agent SDK overview, agent loop, permissions, structured outputs, how Claude Code works, stop reasons, models, and tool use

Part 2 ended with a debt: when I own the runtime, I have to implement trim, a facts block, tiered compress, and placement myself. Claude Code already does a version of that. A product I ship does not get it for free.

This is Part 3 of 6 in my Agentic Coding with Claude Code series. The working model I am keeping is simple: the model reasons, the harness decides what happens next.

Claude Code is one harness. The Agent SDK is that same loop as a library. The Messages API is the loop with the cover off: every response has a stop_reason, and your code chooses whether to run a tool, ask the user, route, escalate, or halt.

Later posts will cover MCP, evals, and guardrails. This one stays on the runtime around the model.

Read More