← All courses

Course

Agent-Ready Go CLIs

Coding agents like Claude already run your Go CLI. Make them good at it: fewer wasted steps, no surprises, and nothing destructive without asking.

Who it's for

Go developers who maintain a CLI and want AI agents like Claude Code to use it safely and well.

Not for: People looking for a first introduction to Go, or to CLI basics — this assumes you already have a CLI.

What you'll build

  • An agent-readiness review of your own CLI
  • Help text and a skill that make agents correct and efficient
  • Permission rules and dry-runs, so nothing destructive runs without asking
  • Evals that prove the agent got better, not just different
  • A clear call on whether an MCP server is worth building

Planned curriculum

This is the plan today. Waitlist feedback will shape the final version.

  1. 01

    Agents already use your CLI

    How coding agents discover and call command-line tools, and where they waste steps or go wrong.

  2. 02

    Reviewing a CLI for agent-readiness

    A practical checklist: output, exit codes, errors, non-interactive use, and the surprises agents trip on.

  3. 03

    Help text and skills that guide agents

    Writing for a reader that takes you literally, so agents pick the right command the first time.

  4. 04

    Fencing commands

    Permission rules and dry-runs, so destructive actions always ask first.

  5. 05

    Proving it with evals

    Measuring agent behavior before and after, so improvements are real.

  6. 06

    When MCP is actually worth it

    What an MCP server adds over a well-built CLI, and when it is not worth the extra moving parts.

Before you start

  • Comfortable writing Go
  • A CLI of your own (or one you maintain) to work on

Your instructor

I'm Marian Montagnino: a senior software engineer, author of Building Modern CLI Applications in Go, international GopherCon speaker, and a founding engineer on Netflix's cloud workstation platform. On my YouTube channel, Build It in Go, I build real things with Go and AI and show the real numbers — what worked, what broke, and what it cost.

Questions

When does it start?

When enough people join the waitlist. Everyone on the list hears first and gets the founding price.

How do I pay?

When enrollment opens, checkout runs through Gumroad. Joining the waitlist is free and commits you to nothing.

Want this for your whole team?

I also run it as a workshop for engineering teams. See workshops.

Want to build the agent itself, not just the tools it uses? Build AI Agents in Go →