Raccoon AI
A general purpose collaborative AI agent
Start new thread
Shubh Saraswat

Raccoon AI - A general purpose collaborative AI agent

by
Raccoon AI
•
Raccoon AI is a collaborative AI agent and workspace for getting real work done. You describe what you need and build it together with an AI agent that has its own computer, terminal, browser, and internet. You see every thought, every file it creates, every decision it makes. You steer when it drifts. You ship when it's right. Deploy web apps. Run deep research. Analyze data. Create pitch decks, videos, images, documents.

Add a comment

Replies

Best
Shubh Saraswat
Maker
đź“Ś
Hey Product Hunt, I'm Shubh, Co-Founder of Raccoon AI. Raccoon AI is like having something between Claude Code and Cursor in the web. The agent has its own computer with a terminal, browser, and internet, and you're sitting right next to it watching it work. You can talk to it mid-task, send it more files while it's still running, or just let it go and come back to a finished result. It's the kind of product where you open it to try one thing and end up spending two hours because you keep thinking of more things to throw at it. You start with "research this market" and thirty minutes later you have a report with charts and real citations, and then you're asking it to turn that into a pitch deck, and then you're editing the slides inline, and then you realize you also need a brand identity so you ask for that too. Same session, everything in one workspace, real files you can download or publish. The modalities are practically unlimited. It connects to Gmail, GitHub, Google Drive, Notion, Outlook, and 40+ other tools. You can add your own via custom MCP servers. Raccoon AI is powered by our in-house built agents SDK, ACE, which is currently SOTA on GAIA benchmark with a score of 92.67, we are releasing the technical report on Hacker News today. It's free to start and you can use code PH5X to get one month of Plus 5x for free. Signup here: https://raccoonai.tech/login. I'll be around all day, happy to answer anything and gather a lot of feedback. Find us on X: https://x.com/raccoonaihq https://x.com/shubh_saras https://x.com/avi_agarwal2001 https://x.com/_pratikpakhale https://x.com/VedantUttam
Rohan Chaubey

@shubh_saras Many congratulations on the launch, Shubh! How does Raccoon handle long running workflows without losing context or crashing?

Shubh Saraswat
Maker
@rohanrecommends Thanks Rohan! Every conversation gets its own sandboxed computer that persists for the entire session. Files, installed packages, environment variables, running processes, everything stays in place no matter how long the workflow runs. For context specifically, the agent uses automatic summarization so the conversation window never hits a wall. It compresses earlier parts of the session while keeping the important details accessible. So if you start with research, then build a report, then turn that into a deck, then edit the deck, it remembers the full chain even if there have been loads of new messages. You can also leave a background process running (like a dev server or a long data pipeline), go do something else in the same session, and come back to it. The sandbox doesn't reset between messages.
Curious Kitty
đź’Ž Pixel perfection
You support 40+ integrations plus custom MCP servers—how do you think about security and governance for tool access (least-privilege scopes, audit logs, permission prompts, secrets handling), and what’s your recommended setup for a team that wants to automate real work without risking accidental actions in sensitive systems?
Shubh Saraswat
Maker

@curiouskitty I love this question, here's how we handle it today:

  • Most connectors use OAuth or API keys with scoped permissions, the agent can only access what you've explicitly granted and you can disconnect a connector anytime, and the access is revoked immediately and all related data is deleted from our systems.

  • On top of that you can choose which tools you want to enable or disable for any particular connector, this way you can choose to give less privileged access to the agent even if the connector's scopes are broader.

  • Everything runs in a sandboxed environment, each session gets it's own isolated computer. The agent's terminal, browser, and file system are containerized. It can't touch anything outside that sandbox unless you've connected a specific integration.

  • Custom MCP servers are user-controlled. You bring your own server, you define what tools it exposes, and you control the endpoint. We don't inject anything into that connection.

  • On audit/visibility: Every action the agent takes is visible in real-time in the session. You can see its thinking, the commands it runs, the files it creates, and the API calls it makes. Session history is preserved so you can rewind and inspect any step. Raccoon AI today is arguably the most transparent AI agent you will find on the internet.

  • What we're still building: granular per-tool permission toggles (so you could say "read from GitHub but don't push"), team-level access policies, and exportable audit logs(we store them, but there is no way to access them on the UI currently). These are on the roadmap as we move more toward team/enterprise use cases.

  • For a team getting started today, I'd recommend: connect any of the integrations you want, but only enable those which are actually need for a given workflow, use custom MCP servers for anything touching sensitive internal systems so you control the boundary, keep write tools disabled for sensitive connectors, and let the agent come to you with why does it need that particular tool and how it will use it and use Plan Mode to review the agent's proposed steps before it executes.


Happy to dig deeper and answer more questions.

garvit jindal

Really impressive launch 🚀
The managed workspace + real-time steering concept makes agent workflows feel much more tangible and production-ready.

How do you see this evolving compared to IDE-native agents — do you expect most execution workflows to move to browser-based agent environments like this?

Shubh Saraswat
Maker

@garvit_jindal Thank you so much :)

Honestly, IDE-native agents are amazing for coding and that experience is hard to match in the browser. We're not trying to replace that. The longer term vision is for Raccoon AI to connect to your remote systems and live where you work, not just in the web.

But we do think the web is where this is all heading. Karpathy said it well yesterday: "humans are moving up and programming at a higher level now. The basic unit of interest isn't one file anymore, it's one agent."

And when the unit of work is an agent doing research, building a site, analyzing data, creating a deck, all in one session, you need something bigger than a code editor. You need a canvas. A higher order execution platform. The web is the natural home for that.

Kate Ramakaieva
đź’ˇ Bright idea

Oh this is interesting for content workflows. Can it pull from Google Drive, write a draft, then format it into a doc - all in one go? And does it remember context between sessions or does each session start fresh?

Congrats on the launch btw!!

Shubh Saraswat
Maker

@kate_ramakaieva Thank you so much :). And yes, it can do all of that in one go. Each session starts afresh currently, we have got mixed reviews on cross session context sharing from users, and we are considering launching it as a toggleable feature really soon.

Denis Akindinov

How does Raccoon AI handle situations where the agent makes a critical or irreversible action, such as deploying code or deleting files, without explicit user confirmation?

Shubh Saraswat
Maker

@mordrag We have a rewind functionality built exactly for this. If the agent makes an error or starts drifting, you can rewind back to any of previous messages and the state of the workspace is restored to what it was at that point including recovering any deleted files.

For things that happen outside the sandbox like you mentioned deploying code, we have implemented guardrails to prevent the agent to do such things without your explicit permissions, and in case that still happens, the agent is smart enough to rollback things safely and cleanly on your request.

the web-app deployment is awesome, any chance I can host it on a custom domain?

Shubh Saraswat
Maker

@raghav_gangwar yes you can!!

Shubham Pratap Singh

Congratulations on the launch 🎉 🎉 !!!

Shubh Saraswat
Maker

@shubham_pratap Thank you so much, do check us out and leave your feedback 🙏.

Tim Clifford

Hey guys, cool launch! What tech did you use to create the motion graphics video? Looks awesome!

Shubh Saraswat
Maker

@tim_clifford3 a mix of @Screen Studio , @Jitter and @Raccoon AI itself. Screen Studio for recording raw footage of the running session, Jitter for the motion components and Raccoon AI for compiling all clips in multiple ways with a variety of sounds and transitions.

Thank you so much :)

Nikita Savchenko

Great tooling, congrats on the launch! Are you competing with Claude actually? (Cowork, code, etc) Or how this product is different?

Shubh Saraswat
Maker

@nikitaeverywhere in some sense yes, I would consider them to be competitors, Claude Cowork is evolving on the same principles that we built Raccoon AI on, but there are a few fundamental differences:

Cowork is a desktop agent tied to your local machine and files. Raccoon AI runs in the cloud with its own computer, so you can access it from anywhere, any device, so you can start something from your phone and come back to it on your desktop, spin up as many concurrent sessions as you want.

Raccoon AI is generally more equipped for specialised use cases and does things Cowork can't: one-click web app deployment, native video and image generation, presentations with a full presenter mode, a canvas for you to edit images inline which will be soon extended for videos.

Claude Code is a different product entirely, and though Raccoon AI excels at async coding tasks, we don't place ourselves in that market primarily.

The end goal is to make Raccoon AI an ultimate agentic workspace that lives where your work lives, kind of like a high level IDE you can access from anywhere.

Mike Sykulski

hi, congrats on your launch!

I see it's general purpose tool, but I wonder are there specific use cases you want to nail during this launch?

I believe it's hard to expect world class output on every possible task

Shubh Saraswat
Maker

@mike_sykulski yes it is, primarily we are looking at knowledge work, web apps and presentations. E2E Image and video generation workflows are close runner ups.

1234
Next
Last