AI Code that Works · Guide

Your guide to the AICTW MCP

Everything you need to get value from your membership — start to finish, in the order you’ll actually do it. No jargon: your coding agent does the technical part. Here’s what you have, how to plug it in, what to say, and where to go if you get stuck.

Step 1 · What you got

What your membership includes

You now have the AICTW MCP— a senior coach that lives at your project’s root and works alongside your coding agent (Claude Code, Codex, Cursor, Gemini, and more). It scores your repo against the AICTW five-phase method and coaches your agent through what to fix, set up, convert, and secure.

It never touches your files itself. It tells your agent what to look at and what edits to make, and your agent is its hands and eyes — so nothing changes without your say-so.

Want the method behind it first? Read the method — the five phases the coach measures your work against.

Step 2 · Getting connected

Plug it into your coding tool

Connecting is a one-time paste. During onboarding you’re shown your personal access token and a copy-in-place snippet for your exact coding tool — Claude Code, Codex CLI, Cursor, Gemini CLI, VS Code (Copilot), or any other client. Pick your tool, copy the snippet into the file it names, and you’re connected.

Your access token is shown once— copy it somewhere safe when you see it. (Lost it later? That’s what recovery is for.)

Step 3 · The commands

What to say, and when

You don’t memorize commands — you talk to your agent normally and it reaches for the right one. But it helps to know the shape of what it can do. The canonical path is short: once you’re connected, run assess for a read-only first look — it scores your repo and hands back a prioritized plan of what to fix first — then run setup to scaffold it the AICTW way (plan-first, so you approve the change-set first). Saying “get me started” gives your agent the same orientation if it needs a nudge.

  • assessread

    Score your repo against the five-phase method and get a prioritized plan of what to improve first. Start here.

  • askread

    Ask in plain language — “how do I…” or “what should I…” — and get grounded guidance and the right playbook to run next.

  • postread

    Turn a messy problem into a clear community question — your goal, stack, what you tried, and the exact question, formatted and ready to paste. You review it and post it yourself; we never post for you.

  • planread

    Plan a feature the AICTW way — a short interview, then a permanent plan file threaded through the five-phase method for your agent to work from.

  • packplan-first

    Make your repo work the AICTW way for the one task you're about to hand off — pack the method steps, this repo's rules, and a verification checklist into a Context Pack under `.aictw/packs/`. Plan-first: you see it before your agent writes anything.

  • composeread

    The hand-off — the exact prompt to paste into your coding agent for one moment (first prompt, implementation, verification, bug-fix, or task-split), grounded in your plan and Context Pack. It writes nothing; you paste it and run it.

  • implementplan-first

    Apply one of our playbooks or briefs to your repo — name it (or describe it) and we resolve the content for your agent to write.

  • fixplan-first

    Remediate an issue from your AICTW report using our playbook for it — pass the fix id from an assess gap or a scan finding.

  • recommendread

    Stack-aware recommendations — what to improve, why it matters, and honest migration advice where a swap is involved.

Session prompt

A prompt to sharpen a session

Beyond the commands, a slash-invocable prompt pressure-tests an idea before you build — run it before plan. Type it in your client the way you’d type any slash command.

  • /ideate

    Pressure-test an idea before you build — validate what you bring, or generate candidates, then hand it to plan. You run the research; the coach never invents numbers.

Client doesn’t show slash prompts? “get me started” does the same job — just say it in plain language and your agent will orient itself.

When you’re ready to build: say plan for a feature, then pack to bundle the method steps, your repo’s rules, and a checklist into a Context Pack — and compose to get the exact prompt to paste into your agent (and again to verify the work when it claims done). All plan-first: you see everything before a line is written.

A couple of commands run under the hood — setting your repo up the AICTW way (a fresh folder or an existing project) and the security review your agent calls for you — so you never have to think about them. For the full method each command measures against, see the method.

Step 4 · Weekly updates

Fresh guidance, every week

The coach ships new teaching and content on a regular cadence — each drop bundles the week’s guidance so your agent can put it to work in your repo. You never fall behind by default: a quick check-in pulls whatever’s newest.

To see the latest: ask your agent “what’s new this week?” — it lists the newest teaching, newest first, and points at the current drop.

To pull one in: tell your agent to run implement with the update it named — that resolves the content for your agent to apply, plan-first. Making a habit of the weekly check-in is the cheapest way to keep a project current.

Step 5 · Recover access

Lost your token? Get a new one

Your token is shown only once when it’s issued, so if you lose it — new laptop, cleared config, or you just never saved it — you don’t email anyone. Head to the recovery page, confirm it’s you with a one-time email link, and a fresh token is reissued to you right there (the old one stops working).

Step 6 · Getting help

Still stuck? We’ve got you

Two good moves when something isn’t working. First, ask the coach itself — say “how do I…” in plain language and it answers with grounded guidance and the right next step. Second, bring it to the community — the AICTW Skool group is where members and the team help each other.

We use cookies

We use cookies and similar technologies to keep this site working, measure how it performs, and (with your permission) personalise ads. Choose your preferences below.