Christophe Pasquier 🇺🇦

Today we're introducing Slite as the first self-maintaining Knowledge Base Your wiki docs go outdated everyday, and your agents depend on them for your most ambitious workflows. Slite monitors your docs' accuracy from all other work tools (slack, jira, linear, codebase, etc) while you sleep and updates them. It even directly talks to your AI agents to give them accurate context without burning their context window or tool calls. You've been trying to build this 'company brain' for months now with hacky MVPs This is the final version of what you need, already built to be collaborative, headless, and secure. Check it out and HMU if you need this for your team, I'll be running personal onboardings for the next few days!

The @slitehq culture in a screenshot. (Andrew is our new support manager)

Very few teams are building a company brain. Most are just wiring together MCPs and hoping scattered docs behave like a system. 6 terms you need to know if you're building one (or evaluating GBrain): 𝟭. 𝗞𝗻𝗼𝘄𝗹𝗲𝗱𝗴𝗲 𝗚𝗿𝗮𝗽𝗵 The layer that maps how company knowledge connects. Linking people, projects, decisions, customers, product areas. It gives AI structure, provides the foundational context AI needs to reason, not just raw text to search through. 𝟮. 𝗠𝗖𝗣 A protocol that helps AI connect to tools and take action across them. Anthropic introduced it; In 2026 it has become the dominant AI integration standard. But a pile of MCP connections does not automatically create shared understanding. Access is not memory. 𝟯. 𝗦𝗸𝗶𝗹𝗹𝘀 The agent’s functional orchestration logic. MCP handles the underlying API connection, Skill defines the higher-level execution steps required to finish a job, such as summarizing a sales call, updating a product specification, or routing user requests. 𝟰. 𝗛𝘆𝗯𝗿𝗶𝗱 𝗦𝗲𝗮𝗿𝗰𝗵 The combo of keyword search and semantic search. One catches the exact phrase, the other catches the meaning. Teams need both, because company language is messy, acronym-heavy, and constantly changing. 𝟱. 𝗔𝗴𝗲𝗻𝘁𝗶𝗰 𝗥𝗔𝗚 RAG with a brain. The agent can route queries to specialized knowledge sources, validate retrieved context, and make dynamic decisions about what information to use. 𝟲. 𝗖𝗼𝗺𝗽𝗮𝗻𝘆 𝗕𝗿𝗮𝗶𝗻 A living system that turns company knowledge into something AI and humans can rely on. Connected, contextual, permission-aware, and continuously updated. This is the difference between AI that sounds smart and AI that is useful at work and always up-to-date. In 2026, we don’t win by adding more AI touchpoints. We win by giving AI a shared brain to work from. Are you building one? Share your expert take below 👇 Which must-know term would you add?

New in Slite Desktop App: Native Tabs 📑 You can now hop between docs without losing your place, reorder your workspace just like in Chrome, or pop tabs out to work side by side. (And yes, your favorite Chrome tab cycling keyboard shortcuts work here too! ⌨️) Try it out in the desktop app → slite.com/download?utm_s…

@simonw Also behind performance a fully misanthropic view of the world

The idea of "AI employees" feels so short-sighted to me - both disrespectful to humans and a complete misunderstanding of what these tools can do and how to best put them to work You may as well start adding Excel spreadsheets to your org chart

@koenbok Landing on an app or website expelling Claude vibes is an instant nogo, makes me instantly doubt any care has been put in it

If your site looks too vibe coded, I’m going to assume your product probably is too.

New Slite website is up! 🛰️

To all saying Notion is becoming bloatware, come back to simplicity 👉 @slitehq

@jittervideo This team is on fire, gg on the launch!

Introducing Superagents ✨ Remix, resize, translate. Tweak small details or animate complex scenes. Create anything you can describe. AI agents, now built right into Jitter 👇

@hey_madni Learning languages has never been about transactional interactions, there is no impact on Duolingo from all these

Duolingo is cooked 💀 GPT-Live fixes grammar while you speak

Here is what happens when Fable is trained in a world pre-Fable

@zodchiii @tanmaigo Self maintenance with HITL is what we offer @slitehq, check it out!

@tanmaigo is there a consolidation step in the agent’s loop? unfiltered memory gets toxic very fast, If every random observation becomes a node that could be bad

We raised $136M to kill Slack. Introducing PromptQL: The first AI version of Slack. Here’s how it works:

Just sumbled upon the 1st app Icon I ever designed!

Is your engineering documentation quietly drifting away from reality? Your code moves, decisions happen in Slack, a pull request changes the implementation, a Linear ticket updates the plan, and the docs just sit there pretending everything is still true. That’s how trust in a developer wiki slowly disappears. A stale doc is worse than a missing one, because it still looks legitimate for humans and agents. @Christophepas walks through how Slite tackles this with a self-maintaining knowledge base, the first one built to detect when documentation has drifted from reality, propose the fix, and route every change through human approval before it becomes truth. Instead of relying on someone to remember a cleanup, the system checks your docs against the places where engineering work actually happens. → 𝗬𝗼𝘂𝗿 𝗰𝗼𝗱𝗲𝗯𝗮𝘀𝗲 When the implementation changes, your docs should not be left behind. Slite watches that gap before it turns into confusion. → 𝗚𝗶𝘁𝗛𝘂𝗯 Technical context often changes in review, long before someone updates the written explanation. → 𝗦𝗹𝗮𝗰𝗸 The answers engineers trust every day are usually buried in threads, not neatly copied into the wiki. → 𝗟𝗶𝗻𝗲𝗮𝗿 Plans evolve, scope changes, edge cases appear. Documentation needs to reflect that motion. The Slite Agent does that across 20+ connected tools, then proposes exact edits next to the original text, so the right owner can review and approve the update in one click. Full walkthrough in the comments 👇 What's the most out-of-date doc you've ever been burned by?

Is your AI good enough to operate parts of the company? The models are not the issue. Your company knowledge is. Refund logic lives in someone's head. Pricing context is buried in a Slack thread from 2024. Half the workflow still depends on Sarah knowing where everything is. Humans can improvise around that kind of chaos. Agents cannot. They need structure. They need current information. They need knowledge they can actually use. So instead of obsessing over which model to plug in, IMHO we should be asking a different set of questions: → How do we turn scattered know-how into something organized? → How do we keep it fresh as the business changes? → How do we make it reliable enough for an agent to act on? Some people call that a 𝗖𝗼𝗺𝗽𝗮𝗻𝘆 𝗕𝗿𝗮𝗶𝗻. And right now, that phrase means something different depending on who you ask. For some teams, it is a wiki. For others, it a memory layer for my agents. For others, it is a continuously updated knowledge base fed by ongoing team activity. That ambiguity is exactly why we made a short survey. We want to understand how teams are keeping knowledge alive today, what is actually working, and where things are still falling apart. We are turning the responses into an interactive ebook, and if we love your answer, we may feature you in it as a contributor. Take the 3 min survey: docs.google.com/forms/d/e/1FAI… If this topic is on your mind too, I would genuinely love to read your take and I'd be happy to have a virtual coffee chat with you too. 🫶

@Timur_Yessenov @slitehq It’s the full promise of Slite, we let you put your docs on self maintenance to fix these contradictions

@slitehq Source, owner, timestamp is the right direction. The missing test: what happens when Slack says one thing, the handbook says another, and CRM is stale? A company brain should return the conflict, not average three sources into one confident answer.

Would you trust your Claude + MCP setup to handle company data while you sleep? We tested it. MCPs and custom tools are great at executing single actions, but they weren't built to reason across your company's full context: → no sense of what's authoritative → no entity resolution → no check before a wrong answer reaches you Slite Agent was built to close this gap: • a query planner breaks down the question • searches every source in parallel • reranks • checks consistency • and returns an answer with source, owner, and timestamp attached • never auto-applies (HITL always) We're writing an ebook on the company brain, and we want it shaped by anyone trying to make sense of their company's data and knowledge (whether you're experimenting with AI agents, or looking to organizing your wikis). Got 4 minutes? Tell us about your setup and we'll quote you: forms.gle/R4dLqEjFxehHkv… Find the link to the deep-dive in the comments below👇

I stopped writing documentation from scratch. And built docs that write and maintain themselves instead. A quick look at my setup in this video: → 𝗚𝗲𝗻𝗲𝗿𝗮𝘁𝗲 𝘁𝗵𝗲 𝗳𝘂𝗹𝗹 𝗱𝗼𝗰 𝘀𝗲𝘁 𝗳𝗿𝗼𝗺 𝘀𝗰𝗿𝗮𝘁𝗰𝗵 The @slitehq agent creates a complete documentation base by pulling context directly from your company knowledge, across all your sources. No need to start with a blank page, chasing scattered context across tabs, or waiting for someone to “get to it later.” → 𝗔𝘀𝘀𝗶𝗴𝗻 𝗼𝘄𝗻𝗲𝗿𝘀 𝘀𝗼 𝗲𝘃𝗲𝗿𝘆 𝗱𝗼𝗰 𝗵𝗮𝘀 𝗮 𝗵𝗼𝗺𝗲 Documentation only works when responsibility is clear. The agent helps assign owners, so every piece of knowledge has someone connected to it instead of floating around as shared team debt. → 𝗩𝗲𝗿𝗶𝗳𝘆 𝗮𝗻𝗱 𝘀𝗲𝗹𝗳-𝗺𝗮𝗶𝗻𝘁𝗮𝗶𝗻 𝗼𝘃𝗲𝗿 𝘁𝗶𝗺𝗲 This is my favorite part. The agent checks each doc against your sources every week, so your (support) team isn’t relying on outdated answers or half-true process notes. It keeps the system alive, not frozen in time. And when documentation maintains itself, your team gets to spend less energy on manual upkeep and more on helping people. Curious to hear from you - who, what or how you are keeping your docs up-to-date? 👀 Learn how to set up your own self-maintaining knowledge base to keep your docs from going stale: slite.com/learn/self-mai…

We’re entering an era of paid context

We're putting together the first comprehensive study on Company brains. I've been talking to dozens of builders and operators implementing one in the last 2 weeks, from the @garrytan's GBrain approach, to @karpathy's LLM wiki style, to simply our own customers @slitehq Now we need to collect the learnings from a much broader audience. Wether you just are just curious about the space or a hardcore builder, share your takes (and speak to me, I put a cal link at the end, we'll feature the best tools, quotes, and stories 🔥) docs.google.com/forms/d/e/1FAI…

We think Europe needs more pro-European, pro-tech, pro-abundance, pro-optimism propaganda. So we made it. Posters, t-shirts, hoodies. Full-on propaganda. We've been handing them out at robotics meetups and clubs all over Europe for weeks, and people love them. Why we are doing this: Europeans know more about the companies in the States than the ones right here at home. We need to change that. Time for propaganda. Every poster features a real European company: 🦾 Allonic is braiding humanoids. A completely new approach to building robotic parts. If it works, a number one company worldwide, right here in Europe. 🚜 Voltrac, the electric autonomous tractor (we made a whole video about them). 🏗️ Automated Architecture (AUAR), doing micro-factories for wooden houses.  🪨 Gondor Industries, a stone mason company using robotics. And many, many more to come. This is the first of many drops. More launching in the next few weeks, including flags, EU posters and more. 👉 Get yours at prototype-shop.com For Europe 🇪🇺🔥 PS: If you're organizing a robotics event or meetup, have a student club or hacker space, comment or message us and we'll send you a swag box.