This product was not featured by Product Hunt yet. It will not be visible on their landing page and won't be ranked (cannot win product of the day regardless of upvotes).
Product upvotes vs the next 3
Waiting for data. Loading
Product comments vs the next 3
Waiting for data. Loading
Product upvote speed vs the next 3
Waiting for data. Loading
Product upvotes and comments
Waiting for data. Loading
Product vs the next 3
Loading
Getzilla
Let AI write code. Keep only what passes the checks.
Getzilla lets your AI coding agent write code, then checks it before it stays: tests run, a different AI reviews it, the author cannot approve itself, and every task gets its own branch, PR and saved evidence. Free, open source, MIT.
Hi Product Hunt! I'm Dmitry, the maker of Getzilla. By day I'm a Head of Growth: for over 15 years I've run paid acquisition and analytics, and I led an international team of 20+ specialists. Lately much of my work is automation: n8n workflows for ad APIs, server-side conversion tracking, an MCP payback calculator. I build all of it with AI coding agents, and that's where Getzilla came from.
Vibe coding is fast and fun: you ask, the agent writes the code, and you hope it works. The problem is that you usually find out later what it broke. The same AI that wrote the change also says "it works", the change sometimes lands straight in main, and the next session forgets everything and repeats the same mistakes.
Getzilla keeps the fast part and adds the checking part. The rule is "vibe first, check second". The agent still builds what you asked for, on its own branch. Before that work is allowed to stay:
- tests and other checks really run, and the results are saved next to the task;
- a different AI reviews the change, and the author cannot approve itself;
- every task gets its own branch and pull request;
- each task leaves a folder with the request, the plan, test results and reviews;
- recurring lessons go into mistakes.md and decisions into decisions.md, so the next session starts with context;
- small changes take a short path, risky ones get more checks and a human sign-off;
- when something is unclear, Getzilla stops instead of guessing.
It runs on your machine, inside your repository. It is not a hosted service. You bring your own agent (Qwen Code by default, or Codex, Claude Code, Gemini CLI, GitHub Copilot CLI, Grok Build) and your own models. There is also an optional EU AI Act evidence layer, off by default, for projects whose AI features reach people in the EU. It produces engineering evidence, not legal advice.
Getzilla is free and open source under the MIT license. Public repositories are gated by GitHub Actions for free. Private repositories can use Trust CI, a separate paid check on dedicated hardware, priced on request.
How to try it (Linux; Windows and macOS commands are in the README). The installer is pinned to the v2.3.0 release and only runs if its SHA-256 matches:
Then take the read-only tour with python3 scripts/getzilla_demo.py --open, or ask for a plan for your own repo with python3 scripts/install_into.py --plan /path/to/your/repo. The plan only reads your project and changes nothing.
What I'd love feedback on:
1. Did the install work on your OS and with your agent? Where did it stop you?
2. Is the "vibe first, check second" flow clear from the README and the tour?
3. Which checks or reviewers would you want added for your stack?
4. Where does it feel too strict, and where not strict enough?
I'll be here all day answering questions. Thank you for taking a look!
About Getzilla on Product Hunt
“Let AI write code. Keep only what passes the checks.”
Getzilla was submitted on Product Hunt and earned 0 upvotes and 1 comments, placing #441 on the daily leaderboard. Getzilla lets your AI coding agent write code, then checks it before it stays: tests run, a different AI reviews it, the author cannot approve itself, and every task gets its own branch, PR and saved evidence. Free, open source, MIT.
On the analytics side, Getzilla competes within Open Source, Developer Tools, Artificial Intelligence and GitHub — topics that collectively have 1.1M followers on Product Hunt. The dashboard above tracks how Getzilla performed against the three products that launched closest to it on the same day.
Who hunted Getzilla?
Getzilla was hunted by Dmitry Vinokurov. A “hunter” on Product Hunt is the community member who submits a product to the platform — uploading the images, the link, and tagging the makers behind it. Hunters typically write the first comment explaining why a product is worth attention, and their followers are notified the moment they post. Around 79% of featured launches on Product Hunt are self-hunted by their makers, but a well-known hunter still acts as a signal of quality to the rest of the community. See the full all-time top hunters leaderboard to discover who is shaping the Product Hunt ecosystem.
For a complete overview of Getzilla including community comment highlights and product details, visit the product overview.
Hi Product Hunt! I'm Dmitry, the maker of Getzilla. By day I'm a Head of Growth: for over 15 years I've run paid acquisition and analytics, and I led an international team of 20+ specialists. Lately much of my work is automation: n8n workflows for ad APIs, server-side conversion tracking, an MCP payback calculator. I build all of it with AI coding agents, and that's where Getzilla came from.
Vibe coding is fast and fun: you ask, the agent writes the code, and you hope it works. The problem is that you usually find out later what it broke. The same AI that wrote the change also says "it works", the change sometimes lands straight in main, and the next session forgets everything and repeats the same mistakes.
Getzilla keeps the fast part and adds the checking part. The rule is "vibe first, check second". The agent still builds what you asked for, on its own branch. Before that work is allowed to stay:
- tests and other checks really run, and the results are saved next to the task;
- a different AI reviews the change, and the author cannot approve itself;
- every task gets its own branch and pull request;
- each task leaves a folder with the request, the plan, test results and reviews;
- recurring lessons go into mistakes.md and decisions into decisions.md, so the next session starts with context;
- small changes take a short path, risky ones get more checks and a human sign-off;
- when something is unclear, Getzilla stops instead of guessing.
It runs on your machine, inside your repository. It is not a hosted service. You bring your own agent (Qwen Code by default, or Codex, Claude Code, Gemini CLI, GitHub Copilot CLI, Grok Build) and your own models. There is also an optional EU AI Act evidence layer, off by default, for projects whose AI features reach people in the EU. It produces engineering evidence, not legal advice.
Getzilla is free and open source under the MIT license. Public repositories are gated by GitHub Actions for free. Private repositories can use Trust CI, a separate paid check on dedicated hardware, priced on request.
How to try it (Linux; Windows and macOS commands are in the README). The installer is pinned to the v2.3.0 release and only runs if its SHA-256 matches:
Then take the read-only tour with python3 scripts/getzilla_demo.py --open, or ask for a plan for your own repo with python3 scripts/install_into.py --plan /path/to/your/repo. The plan only reads your project and changes nothing.
What I'd love feedback on:
1. Did the install work on your OS and with your agent? Where did it stop you?
2. Is the "vibe first, check second" flow clear from the README and the tour?
3. Which checks or reviewers would you want added for your stack?
4. Where does it feel too strict, and where not strict enough?
I'll be here all day answering questions. Thank you for taking a look!