root

root is a private, encrypted & local AI harness. Works just like Cursor or Codex, but yours without limits.

Built by @palmaierc

[root @rootsh]

turn the lights on

https://t.co/aBbJDBOo16

[PaLM AI - $PALM @palmaierc]

What if you could build an entire videogame with a single prompt, with different MCPs working together in the background and using all the tools to combine created .obj models on one thread, while incorporating them into the code on another thread?

You can do this, and much more, with @rootsh. The two model combinations below were completely made with the new and advanced MCP framework for @Blender that ships as a default plugin for root.

New update v0.0.3 drops today. If you are willing to test it out and have not yet, join https://t.co/pVHlIJgV0W and get a tester key that's valid for 3 months.

Note: the idea is not as much making games but rather can root survive creative tasks done in iterative loops. If it can make a strategy game with 3D models and logic in one prompt, it can do anything.

Welcome to utilizing frontier-level AI while actually owning the results and retaining your privacy.

[root @rootsh]

privacy is a codeword for crime

[root @rootsh]

brrrrr

[PaLM AI - $PALM @palmaierc]

root v0.0.4 has features like never before seen in a local, privacy-focused AI setup.

Nothing in the video is done by hand, or without much thought put into it. @rootsh has done all the work as demonstrated.

Feeling lazy, drunk vibe coding at 4AM, or just had a long day and your prompts reflect that? No worries - toggle on auto-prompting on https://t.co/3s0MYGBnBF and allow the AI to think your prompts for you.

The one-prompt pipeline ensures that tasks are handled effectively, with all of the UI from the chat itself to the file system working together to keep you in the loop - that is, if you are interested.

root automatically recognizes all application formats and allows you to one-click launch an in-app terminal that reports directly to the chat context.

The application flow can automatically develop 3D models and handle context requests up to 1M tokens with auto-compacting post-task.

This allows for direct integration of freshly built visuals to the applications at hand, which makes it insanely efficient for game development.

Is it still buggy? Yes. Is it getting there? Yes. We started five days ago. Some have been building for years and months. Will we get there? No doubt.

A note for our X Community:
The distribution wallet has been filled with a record-amount of $PALM: https://t.co/hMXgejoQqD

[root @rootsh]

@BaronVonHolland root

[root @rootsh]

Find and pull a model. Install the tools to effectively use it. Multi-task with CLIs and IDEs. Learn and build everything from scratch.

Or just use a shortcut. https://t.co/aBbJDBOo16

[root @rootsh]

@chivopost si señor

[root @rootsh]

https://t.co/aBbJDBOo16

[PaLM AI - $PALM @palmaierc]

Development Update 🌴

The Latest Stable Version of root now has automatic updates. This means simply installing the application once, and receiving all updates without navigating back to any downloads page.

For telemetry reasons, you can opt out of the automated updates and download the latest stable versions always from https://t.co/3s0MYGBnBF.

Many more updates to come tonight, including on our Linux version of root, the first frontier-level AI harness app to be released on Linux-based distributions!

[root @rootsh]

it is your duty to do this privately

https://t.co/aBbJDBOo16

[Bittex @BittexXBT]

It is your duty to grind as much as possible on a public ledger in your 20s so you can spend your life in prison

[PaLM AI - $PALM @palmaierc]

🏆 $PALM has reached a milestone of over 12% of the Total Circulating Supply staked, amounting to over 9.26M tokens.

The new week will climax in the distribution of our 1st Protocol Owners Rewards along with a Loyalty Program +% for those that matched the criteria before July 1st.

It is never too late - stake your $PALM tokens at https://t.co/NZk68Naghy to make your bag work for you, instead of you working for it.

Thank you massively for the continued trust & support 🙏

[root @rootsh]

@perserpersian privacy is a right

[root @rootsh]

@BenjaminMilesUK speed is key

[root @rootsh]

@omgsidewalks paying @AnthropicAI & @OpenAI for usage

[PaLM AI - $PALM @palmaierc]

/root Public Release v 0.1 is live.

Time to reclaim your privacy.

@rootsh is a frontier-level harness for your local AI. It can do everything @claudeai, @OpenAI or @cursor_ai can - but with no usage limits and full privacy, including zero telemetry transmitted outside of the encrypted /root files.

Privacy and usage both begin at the root level. Start today. Comes with in-built documentation and a full start-up guide that takes you through the app like no other platform.

Get it here 👉 https://t.co/3s0MYGBVrd

[root @rootsh]

@Spectre__AI @web3kravchuk better set one up with root

[PaLM AI - $PALM @palmaierc]

Development Update ⚒️

The $PALM vision is back, and we are making it happen.

/root is becoming a local AI layer for your PC. Early Access has been released to a select group of $PALM holders and will expand soon. A full video breakdown of all features is coming – meanwhile, here are some highlights from this week of development!

Instead of being confined to a single app, /root can run across the OS through the app miniframe. Once active, it can see what is happening on your screen and help complete tasks by clicking, running commands, creating files, managing processes, and coordinating with background agents.

Because /root has deep local access, visibility and control are core requirements. The telemetry page records every local and internet data share, inbound and outbound, so privacy is something users can inspect instead of simply trust. The app will also monitor network activity and alert users about possible intrusions, new WiFi connections, or suspicious scraping.

The next layer is a lightweight local assistant that sits on your screen, can be dragged anywhere, and runs on its own dedicated thread. It helps notice what you are doing, supports thought processing, keeps you focused, and passes reasoning notes to background tasks. It can also be adapted into an accessibility, gaming, grammar, or workflow assistant using a sub-2GB model with voice included.

🌴 More updates are on our Telegram and in the Spaces tonight at 7PM UTC!

[PaLM AI - $PALM @palmaierc]

Development Update 🌴

/root is nearing the first Public Release promised for later this week. A select group of holders already has access to it through Early Access, selected through activity during Spaces.

We've recently implemented lightweight terminals and a non-shell coding assistant that is 50% faster than previously. This allows delegating simple code runs to more lighter models.

For our $PALM stakers — today on the 30th you will see your Staking Dashboard w/ the token amounts again along with the new rules and our updates.

/root also features a level of customization not seen in pre-built, plug&play AI workspace products. You can simply toggle on/off a feature from the Configuration panel located at the Settings tab.

More to come — join our Telegram for almost hourly updates & discussion!

👉 https://t.co/pVHlIJgnbo

[PaLM AI - $PALM @palmaierc]

1 prompt, 2 commits.

Prior to the first Public Release, we have added full @github (not just Git) support to /root. It can create and manage repositories for you, ensuring that your work is never lost and has commit backups, branches and forks that you can revert to if needed - or tell root to do so.

In comparison to other platforms, root can do absolutely everything in your GitHub session and you need zero past experience (or even knowing what it is) to employ it. Simply ask /root to build a cloud repository for your project, and it will guide you there.

For privacy focused users, you can set up your own repository using the Git plugin through SSH/TLS.

[PaLM AI - $PALM @palmaierc]

Work-in-Progress: a fully local AI modelling done through /root with $0 cost.

Simply having a coding agent and a conversational assistant isn't enough. A full-stack AI application running at the OS levels should encompass many fields, which is exactly what we are doing.

Using a tool side-by-side with the two applications open can be a pain, especially for our target audience (standard customer laptop). To fix this, there is a dedicated "Follow Me" assistant to Blender that can both see whats happening and take instructions and report on progress.

This is the very early alpha, a more refined version will be included in the 1st public release.