Osaurus

Swift-native, fully offline, and open source is a rare combo — most agent harnesses assume you're fine piping everything to the cloud. The cryptographic identity piece is intriguing; I'd love to hear more about how persistent memory works across sessions and whether it's local-only storage. This feels built for people who actually care about owning their stack. Nice work.

Congrats on the launch. The approval gate feels like the right trust boundary for an agent that can touch files, Calendar, Contacts, and iMessage. Iâm curious how you think about permissions once someone has several agents or projects: are you aiming for per-action approvals only, or eventually a project-scoped policy where an agent can access one context but not another?

this is one of the more compelling "local AI agent" pitches I've seen, the approval gate before every action is the part that actually matters to me, most of these tools just yolo the agent loose on your filesystem. question on the self-scheduling piece you mentioned - does that still fire if the Mac is asleep or the lid is closed, or does it need to actively be awake/plugged in for scheduled runs to trigger?

the approval gate plus "always allow" per action is the right default, but the part I'm actually curious about is the open source contribution angle - if community members start shipping shareable agent skills/plugins that get access to iMessage or Contacts, is there any review step before those get trusted, or is it on the user to audit what a downloaded skill can touch before installing it

Does it take a lot of space in my computer?

Local-first agents is the direction I keep hoping wins. The cryptographic identity part is interesting — most agent frameworks treat identity as an afterthought. Curious how heavy it gets memory-wise with a couple of agents running on an M-series Air?

Just discovered @Osaurus on XUsed my ChatGPT Pro subRecognised my local MLX models from LM Studio💜 Dino avatars

Nice. If Qwen (or some of its models) are still open-source, which is my guess, I can confirm it works fine on a Mac Mini M3 for doing automation tasks on websites, DOM manipulation...

Downloading to check this out now! Just the branding of the dinosaurs is unique enough to pique my interest, would love to know where you got the idea to use them. Make the internet fun again!

How good it the macuse skill? Can the mac be controlled via local LLM? or does that require better models only available via cloud? If local LLM is possible with mac-use skill, what requirment would support local mac-use skill? Btw looooooove the Laundromat Afternoon theme 😸

The approval gate is nice. One action, easy call. Wondering once an agent chains 15 steps to build a deck or deploy a site, does every step stop and wait for me? Fantastic launch!

@tpae Huge congrats on bringing Osaurus to PH! 🎉For users running hybrid workflows (e.g., small local model like Gemma for simple file/routing tasks, but switching to BYOK Claude/OpenAI for heavy reasoning), can an agent dynamically route tasks between local and cloud models within the same workflow, or is an agent fixed to one specific model backend?

honestly this looks great for mac users, the offline angle is a big plus. one thing i'd love to see is a simple visual diff or timeline for what the agent changed locally, like a little activity log you can scrub through. would make it way easier to trust what it's doing in the background

Congrats on the launch! I like that it runs locally and asks for approval before each action. For anything touching Calendar, Contacts, and iMessage, that's the combination I'd want before granting access.Curious how you handle memory and context across sessions while keeping it all on-device.

Finally a Swift-native agent harness that doesn't feel like a wrapper around a Python script. The persistent memory across sessions actually held context from yesterday, which surprised me.

@Osaurus Congrats on the Launch. The Osaurus dino is super cute. I have been waiting to try locally run models and see the performance. What the hardware requirement? is a M4 Mac mini 16gb enough? Eager to try it.

Hi, this is cool. What are some exciting and unique ways that people are using Osaurus and is there a roadmap?

So stoked to unveil @Osaurus to the Product Hunt community!I've been a long-time advocate of open source (going back to the launch of @Mozilla Firefox in 2004!).I've followed products like @Ollama and @LM Studio but neither is Mac-native nor have the character of Dinoki, the little guy I originally fell in love with, long before Petdex existed:While I'm a big Anthropic and OpenAI user, I also believe strongly that local, open source AI is the future of computing, and Osaurus is the user-friendly app bringing this to everyone.Osaurus is like the Firefox of open source AI!You can download and run the latest models from @Z.ai @Qwen @Gemini (Gemma), or BYOK.Now's an awesome time to give Osaurus a try, and if you're feeling bold — the beauty of open source means we'd love your contributions!

Hey Product Hunt, I'm Terence, founder of Osaurus. I spent 20 years shipping software at Netflix, Tesla, and Zillow before going all in on this.Own your AI. Osaurus is an open source AI agent platform that runs on your Mac.MIT licensed, native Swift for Apple Silicon. No account, no subscription, no Electron.We started as Dinoki, a 5MB desktop dinosaur, and grew into a full agent platform through open source community feedback. Now at 7K+ GitHub stars and 175K+ downloads — all by word of mouth.The problem: most AI assistants run on someone else's server.Your files, memory, and context leave your machine every time you use them, usually behind a monthly subscription.Osaurus keeps all of it on your Mac.What that means in practice:Agents read your Calendar, look up Contacts, send iMessages, and work with your real filesAn isolated sandbox writes and runs code, producing real files on your desktop: images, PDFs, presentationsEvery action goes through an approval gate. You see what the agent wants to do before it does itLocal models run fully offline. We built our own Swift MLX runtime, no Python underneathWant a frontier model? Bring your own key (OpenAI, Anthropic, Gemini) or use pay-as-you-go Osaurus Cloud. Either way, your memory and context stay localIt's free and always will be.16GB of RAM gets you started with local models, 24GB recommended.Try it at osaurus.ai. Download takes seconds, no account needed.I'll be here all day answering questions. My favorite Osaurus hack is to set up Agent DB and self-scheduling to create a personal dashboard!