Haven't taken the check yet? Do that first — 14 questions, about four minutes. It scores you on the three areas below and tells you which one to fix first. Already have your result? Jump to your weakest area: Workflow · Guardrails · Fit & spread.

01The model

Three things to get right. A slick workflow with weak guardrails ships fast, unreviewed risk; strong guardrails with no workflow just add friction nobody uses. The rule underneath all of it: the person who prompted it owns the diff.

01Workflow — the agent set up, repos giving it context, people knowing when to reach for it and how to redirect it.
02Guardrails — human review before merge, the same test bar as hand-written code, clear diff ownership, scope control.
03Fit & spread — more than one person capable, a safe path for non-technical people, a shared view of what it’s for.

The check scores each and gives a combined grade from Experimenting to Fluent. The engagement, in a sentence: find the weak one and reinforce it first.

02Workflow

Workflow is whether directing the agent is a repeatable practice or one person's party trick. The test: could a second person start a session and get useful work out without re-explaining the codebase?

What good looks like
  • A coding agent is actually set up and in use — by more than one person, not one champion.
  • Repos give the agent the context it needs — conventions, how to run tests, what not to touch.
  • People know when to reach for the agent and when to just write the code.
  • Prompts and sessions are treated as a skill people get better at — patterns get shared.
  • When the agent goes down a wrong path, people know how to stop, reset and redirect — not start over.
Common failure
  • The lone champion — it works, but only when one person drives it.
  • Context starvation — no agent-facing docs, so every session re-explains the codebase.
  • Start-over loop — when the agent drifts, people scrap the session instead of redirecting.
Your first move
If you're just starting

Get the basics in place: the agent installed and used by more than one person, and an agent-facing doc in each repo (conventions, how to run tests, what not to touch). Then run a short session on when to reach for it — and when not to.

If it's partly there

The agent is used but the workflow is folk knowledge. Make "share what works" deliberate — a place for prompts and patterns — and teach the skill of stopping and redirecting the agent when it drifts, rather than starting over.

If it's already strong

Keep the agent-facing docs current as the codebase moves, keep the shared patterns alive, and keep teaching new people the judgement of when to prompt versus when to type.

Go deeper

  • Add an agent-facing doc to each active repo: conventions, how to run tests, directories that are off-limits.
  • Start a shared page of prompt patterns that worked, pasted from real sessions with a note on what they were for.
  • Teach one redirect technique — stop, summarise where it went wrong, point at the constraint — instead of discarding the session.

03Guardrails

This is the risky one. Guardrails are what stop fast, confident output from becoming unreviewed risk in the codebase. They are non-negotiable, not optional.

What good looks like
  • Every AI-generated change is reviewed by a human who understands it before it merges.
  • AI-generated code passes the same tests — and ideally new ones — as hand-written code.
  • The person who prompted it owns the diff — its correctness, its security, its fit.
  • Scope control stops the agent touching secrets, prod config, unrelated files.
  • AI-generated code gets a security pass — injection, secrets, unsafe dependencies — as standard.
Common failure
  • Merge-and-hope — AI-written PRs get a lighter look than hand-written ones.
  • Scope creep — the agent edits files, config and secrets it was never meant to touch.
  • No security pass — injection and secret-leak checks only happen for “sensitive” code.
Your first move
If you're just starting

Non-negotiables first: every AI-generated change reviewed by a human who understands it, the same test bar as hand-written code, and "the person who prompted it owns the diff" — its correctness and its security. Add scope control so the agent can't wander into secrets or prod config.

If it's partly there

Review and testing happen but aren't consistent. Make AI-written PRs get extra scrutiny, make test coverage part of "done" for agent work, and add a standard security pass — injection, secrets, unsafe dependencies — not just for sensitive areas.

If it's already strong

Keep the ownership rule enforced in review as speed tempts shortcuts, keep the security check standard, and keep scope control tight as more people use the agent.

Go deeper

  • Write the rule down: “you own the diff you prompted” — correctness, security, fit — and enforce it in review, not aspiration.
  • Add “tests included” to the definition of done for agent-written work, same bar as anything else.
  • Make a standard security checklist part of review for every AI-generated PR: injection, secrets, dependency risk.

04Fit & spread

Fit is whether this is a team capability or one leaver away from gone. The test: is there a shared, written view of what agentic coding is good for here — and what it isn't?

What good looks like
  • More than one person can do it — it doesn't leave if they leave.
  • Non-technical people have a safe path for the right things — internal tools, prototypes, scripts.
  • Agentic coding sits in the real delivery process — branching, review, CI — not a side experiment.
  • There's a shared view of what it's good for here — and what it isn't.
Common failure
  • One-person capability — it leaves if they leave.
  • Side experiment — it never joins the real branch-review-CI process, so it never matures.
  • No shared position — every person has their own idea of what it’s for.
Your first move
If you're just starting

Right now this leaves if one person leaves. Get a second and third person capable, write down what the team has learned, and pull agentic coding into the real process — branching, review, CI — rather than a side channel.

If it's partly there

It's spreading but not yet dependable. Open a safe path for non-technical people to use it for the right things (internal tools, prototypes) with guardrails, and agree an explicit position on what agentic coding is good for here and what it isn't.

If it's already strong

Keep the shared "what it's good for" position revisited as the tools change, keep the non-technical path supported, and keep it inside the normal delivery process, not a fast lane around it.

Go deeper

  • Pair-build on the live codebase with a second and third person until they can run a session solo.
  • Write a half-page “what we use the agent for / what we don’t” and revisit it each quarter.
  • Give non-technical people a bounded lane — internal tools, prototypes, scripts — with the same review rule.

05Do this in the next two weeks

  • Take (or retake) the readiness check and note which of workflow, guardrails or fit scored lowest.
  • For workflow: add an agent-facing doc to each active repo, and start a shared prompt-patterns page.
  • For guardrails: write down "you own the diff you prompted", and add a security checklist to review for AI PRs.
  • For fit: pair-build with a second person, and draft the half-page "what we use it for / what we don’t".
  • Make "tests included" part of done for agent-written work.
  • Pick the single weakest area, put a named owner and a date on its first move, leave the other two as "next".

Still stuck?

If the weak area won't move — review keeps going light on AI PRs, it stays stuck with one person, it never joins the real process — that's the point to bring Jon in. Set up the workflow (agent installed, agent-facing docs, shared patterns), put the guardrails in (review rule, test bar, diff ownership, scope control, security pass), then spread it: pair-building on the live codebase until it lives in normal delivery.