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
Myrmidon
One agent. A whole crew. Every step accountable.
Local-first, open-source agent harness: chat with one lead that answers, codes, browses, or spins up a crew of AI agents. Every step logged, sandboxed, and reviewable
Hey Product Hunt! 👋 I’m Robin, the creator of Myrmidon.
Myrmidon is a free, open-source AI agent platform for your PC. One agent, a whole crew, and every step accountable.
I built Myrmidon because I wanted an AI agent that could do more than answer questions or write code. I wanted one that could actually get things done: investigate problems, use a real browser, work with apps on my Mac, and bring in a team of specialist agents when a task gets complicated.
The idea is simple: you give Myrmidon a goal, and it figures out how to get the work done.
With Myrmidon, you can:
Delegate complex tasks to AI agents that plan, collaborate, and review each other’s work.
Code in your project with checkpoints, isolated Git worktrees, and reviewable diffs.
Automate a real browser and use approved apps on your Mac.
Chat with your agent from messaging apps and continue the same work across channels.
Choose from supported AI models, including signing in with subscriptions you already have.
Keep control through permissions, approval gates, an audit trail, and the ability to rewind changes.
For example, imagine discovering a bug in your checkout flow. Ask Myrmidon to investigate it, reproduce the issue in a browser, delegate the fix to a coding agent, and run the tests. You can follow the work, inspect the changes, and review the result instead of blindly trusting the final answer.
Myrmidon is open source under Apache 2.0, and I’m building it in public. It’s still in beta, and there’s plenty left to improve. I’d love to build this with people who care about capable AI agents that remain transparent and under your control.
The app currently targets macOS on Apple silicon. You’ll need access to a supported AI provider or API key, and provider usage limits and fees may apply.
I’d love your feedback, especially on setup, the first task you try, and where the agent or multi-agent workflow falls short.
What would you delegate to a crew of AI agents if you could watch and verify their work?
About Myrmidon on Product Hunt
“One agent. A whole crew. Every step accountable.”
Myrmidon was submitted on Product Hunt and earned 0 upvotes and 1 comments, placing #159 on the daily leaderboard. Local-first, open-source agent harness: chat with one lead that answers, codes, browses, or spins up a crew of AI agents. Every step logged, sandboxed, and reviewable
On the analytics side, Myrmidon competes within Open Source, Artificial Intelligence and Vibe coding — topics that collectively have 551.4k followers on Product Hunt. The dashboard above tracks how Myrmidon performed against the three products that launched closest to it on the same day.
Who hunted Myrmidon?
Myrmidon was hunted by Michael Robin K. 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 Myrmidon including community comment highlights and product details, visit the product overview.
Michael Robin K
Hey Product Hunt! 👋 I’m Robin, the creator of Myrmidon.
Myrmidon is a free, open-source AI agent platform for your PC. One agent, a whole crew, and every step accountable.
I built Myrmidon because I wanted an AI agent that could do more than answer questions or write code. I wanted one that could actually get things done: investigate problems, use a real browser, work with apps on my Mac, and bring in a team of specialist agents when a task gets complicated.
The idea is simple: you give Myrmidon a goal, and it figures out how to get the work done.
With Myrmidon, you can:
Delegate complex tasks to AI agents that plan, collaborate, and review each other’s work.
Code in your project with checkpoints, isolated Git worktrees, and reviewable diffs.
Automate a real browser and use approved apps on your Mac.
Chat with your agent from messaging apps and continue the same work across channels.
Choose from supported AI models, including signing in with subscriptions you already have.
Keep control through permissions, approval gates, an audit trail, and the ability to rewind changes.
For example, imagine discovering a bug in your checkout flow. Ask Myrmidon to investigate it, reproduce the issue in a browser, delegate the fix to a coding agent, and run the tests. You can follow the work, inspect the changes, and review the result instead of blindly trusting the final answer.
Myrmidon is open source under Apache 2.0, and I’m building it in public. It’s still in beta, and there’s plenty left to improve. I’d love to build this with people who care about capable AI agents that remain transparent and under your control.
The app currently targets macOS on Apple silicon. You’ll need access to a supported AI provider or API key, and provider usage limits and fees may apply.
🚀 Try Myrmidon: https://github.com/Robinbinu/myrmidon
💻 Source code: https://github.com/Robinbinu/myrmidon
🎬 Watch the demo: https://youtu.be/FnOd7sCxWWM
I’d love your feedback, especially on setup, the first task you try, and where the agent or multi-agent workflow falls short.
What would you delegate to a crew of AI agents if you could watch and verify their work?