You're already using AI to write Flutter code. Now make it write code you trust. ACT is a structured workflow from requirements to verified implementation — where problems are caught early, not in the code review.
Testimonial from Holger Matthies: I'm loving ACT. It's accelerating me and giving me more confidence, more control . Repeated review runs catch different critical findings, and that helps me turn a rough spec into something solid before implementation starts. The walkthrough reassured me that I'm using ACT as intended , and that alone made it worth the time. Thank you for ACT, it's truly making a difference .
Holger Matthies
Flutter developer
Testimonial from Jason Holt: ACT is a pretty thoughtfully designed toolkit. It's been really fantastic for me. I've used/tried a few of these different things (superpoweres, planning with files, the vgv AI toolkits) but the design of your system is definitely the most thoughtful and complete . It works really really well together without me having to chase down a new skill everytime.
Jason Holt
Founder of Jason Holt Digital
AI agents are great at following patterns and writing lots of code — when you give them the right structure. Without it, they infer missing requirements, make product decisions, and guess at your conventions, often going off-track. AI is a multiplier — of whatever process you give it. The speed is real. The confidence isn't.
AI follows local context. When conventions and decisions are not documented, patterns drift as features pile up.
The AI produces something, but you can't trust it without reading every line. That's not faster — it's just different work.
AI agents optimize for completion. They will chase green checks with brittle, shallow, or disabled tests that fail to protect real behavior.
Without a proper spec, the AI fills in the blanks with guesses. And you don't discover the wrong assumptions until the PR review.
AI agents write code that compiles. They don't write code that follows Flutter conventions — widget composition, proper state management, theme extensions, idiomatic patterns. You end up fighting the agents or rewriting by hand.
You solved this exact problem last week. But the context is gone and the AI has no memory of what worked. Without a way to capture and reuse insights, every session reinvents the wheel.
You describe what you want to build. The AI asks questions (rather than guessing) and writes a detailed specification, then a second pass tears it apart — finding gaps and wrong assumptions before a single line of code is written.
Then it breaks the spec into focused tasks and implements one bounded piece at a time. Tests lock in behavior and keep future runs aligned.
What if your AI agent actually worked this way?
Introducing the
Agentic Coding Toolkit (ACT) is a collection of commands, agents, skills, and a Flutter knowledge base that turns AI-assisted development into a structured workflow from spec to PR.
ACT is fully open-source readable, installs in two commands, and works with any Flutter or Dart project without changing your project structure.
ACT supports Claude Code, OpenCode, Codex, GitHub Copilot CLI, Cursor, and Antigravity. You'll need one of these installed, along with your preferred LLM subscription or API key.
Watch ACT take a feature request through interview, spec, task breakdown, and execution, with each step verified along the way.
This walkthrough shows the core ACT idea: clarify requirements, write a spec, break work into tasks, and implement it. Some commands and details have changed; use the Quickstart for the current workflow.
From requirements to verified implementation with guardrails at every step
You describe the feature. The AI asks clarifying questions and helps you iterate until all user flows, edge cases, and error scenarios are covered.
The answers become a traceable spec and decision record, with requirements you can review before any implementation starts.
A second pass roasts the spec — looking for gaps, wrong assumptions, and misalignment with your existing codebase.
Turn the spec into focused implementation tasks with acceptance criteria, dependencies, and the context needed for handoff.
The AI implements one focused task at a time, running the relevant checks and reporting what changed.
Six pillars that transform AI-assisted Flutter development from chaotic to predictable
AI works from detailed specs, not vague prompts. Each spec includes user flows, edge cases, and error scenarios — then gets reviewed for gaps before implementation starts.
A repeatable cycle where each stage produces artifacts that feed the next. Interview to spec to implementation tasks — with guardrails at every step.
Tests are baked into implementation so behavior stays explicit and future agent runs stay on track. TDD-style guidance when it helps — without turning every task into ceremony.
Encode user journeys as robot-driven widget tests with stable selectors and deterministic test seams. Real user flows, not isolated unit checks.
10 principles. 15+ patterns. Breaking changes docs. Setup recipes. Available when specs are refined, tasks are created, and code is implemented — not as an afterthought, but as the foundation.
Conventional commits, automatic PR creation with smart descriptions, and full git worktree support for isolated parallel feature development.
Everything included in the full toolkit, organized by skill type, supporting agents, and Flutter knowledge base.
| Skill | What it does |
|---|---|
/act-config | Configure ACT workflow storage and project setup once before using the workflow |
/act-interview | Ask focused questions before requirements are written, so the agent does not guess |
/act-create-spec | Create specs with user flows, edge cases, acceptance criteria, and a record of key decisions |
/act-refine-spec | Adversarially review specs for gaps, wrong assumptions, and codebase misalignment |
/act-create-issues | Turn specs into focused implementation tasks with clear acceptance criteria |
/act-implement | Implement one focused task at a time with context, testing, and verification |
You're past tab-autocomplete. You've used tools like Claude Code and Codex to build real features, and you've seen what happens when the AI works without guardrails. Architecture drifts. Tests pass for the wrong reasons. Every session starts from scratch.
You don't want to stop using AI. You want it to work your way — following your patterns, writing proper specs, and producing code you'd actually approve in a PR review.
Whether you're working solo or on a team, ACT gives you that structure.
This isn't for vibe coders. It's for engineers who care about what they ship.
Start with Lite for the core workflow. Upgrade to Pro for Flutter/Dart expertise, Git workflow support, background agents, setup recipes, testing guidance, and the full toolkit.
Experience how Spec-driven development changes the way you build with AI.
Includes:
Best for AI coding on real projects that require testable and maintainable code.
Includes full toolkit access and updates while your subscription is active:
Best value if you want long-term updates with no renewals. One purchase, lifetime access.
Includes lifetime access to the full ACT toolkit and all future updates:
I've put a lot of care into making this toolkit genuinely useful and I'm confident you'll find it valuable.
But the only way to know if it's right for you is to try it. Get the toolkit, run the workflow on a real feature, and see the difference for yourself.
If you're not happy with it, for any reason, reach out via the contact page within 30 days and I'll issue a full refund within 24 hours.
About the Author
I'm a Flutter GDE and I've been writing code professionally for over 15 years. With my Flutter tutorials and courses, I've helped thousands of developers become better at building Flutter apps over at Code with Andrea.
The Agentic Coding Toolkit is the system I use daily to build Flutter apps with AI assistance — and now it's available to you.
Give your AI agent the guardrails, Flutter expertise, and structured workflow it needs to produce code you trust.