Gul Saeed Khattak

𝗢𝗽𝗲𝗻 𝗛𝗶𝗴𝗴𝘀𝗳𝗶𝗲𝗹𝗱 𝗜𝘀 𝗮𝗻 𝗢𝗽𝗲𝗻 𝗙𝗿𝗼𝗻𝘁 𝗘𝗻𝗱, 𝗡𝗼𝘁 𝗢𝗽𝗲𝗻 𝗠𝗼𝗱𝗲𝗹𝘀 Open Higgsfield AI is getting shared as a free replacement for Higgsfield. It's worth being clear about which layer is actually open. There are roughly three layers in an AI video product. The models that do the generating. The API that hosts and bills for them. And the app you click around in. 𝗪𝗵𝗮𝘁'𝘀 𝗼𝗽𝗲𝗻 𝗵𝗲𝗿𝗲 𝗶𝘀 𝘁𝗵𝗲 𝘁𝗼𝗽 𝗹𝗮𝘆𝗲𝗿 𝗼𝗻𝗹𝘆. The studio UI is MIT licensed and you can run it locally. The middle layer is Muapi, a paid pay-as-you-go API, and every request goes through it with your key. The bottom layer is mostly closed models like Kling, Veo, Sora and Midjourney. I think of it like an open-source shop front for a wholesaler. You can rebuild the shop however you like, but you're still buying stock from the same supplier. That's not a bad thing. A good UI over one API that reaches many models is genuinely useful, and it lets you skip a monthly plan. But "self-hosted, your data stays on your machine" is a stretch when prompts and uploads leave for Muapi's servers. Links to repo 👇 github.com/Autom8AI/Open-Hig…

𝗔 𝗟𝗶𝘃𝗲 𝗠𝗮𝗽 𝗼𝗳 𝗪𝗵𝗮𝘁 𝗖𝗹𝗮𝘂𝗱𝗲 𝗖𝗼𝗱𝗲 𝗜𝘀 𝗗𝗼𝗶𝗻𝗴 You give Claude Code a task and the terminal fills with lines. Read, Grep, Bash, Edit, a subagent, another Bash. By the time you've scrolled back, it's done ten more things. lot-map turns that scroll into a picture. It's a small mod that opens 𝗮 𝗹𝗶𝘃𝗲 𝗺𝗮𝗽 𝗶𝗻 𝗮 𝗽𝗮𝗻𝗲 𝗯𝗲𝘀𝗶𝗱𝗲 𝘁𝗵𝗲 𝗰𝗵𝗮𝘁. In the middle is the agent. Around it are stations: You, Model, Memory, Files, Shell, Network, Browser, Agents, Skills and Hooks. Every time Claude does something, a pulse flies from the centre to the right station. • 𝗥𝗲𝗮𝗱𝗶𝗻𝗴 𝗼𝗿 𝗲𝗱𝗶𝘁𝗶𝗻𝗴 𝗰𝗼𝗱𝗲 lights up Files • 𝗥𝘂𝗻𝗻𝗶𝗻𝗴 𝗮 𝗰𝗼𝗺𝗺𝗮𝗻𝗱 lights up Shell • 𝗙𝗲𝘁𝗰𝗵𝗶𝗻𝗴 𝗮 𝗽𝗮𝗴𝗲 lights up Network • 𝗦𝘂𝗯𝗮𝗴𝗲𝗻𝘁𝘀 show up as small dots circling the Agents station It's not the first agent dashboard. Watching what a coding agent does has become its own category: • 𝗱𝗶𝘀𝗹𝗲𝗿'𝘀 𝗺𝘂𝗹𝘁𝗶-𝗮𝗴𝗲𝗻𝘁 𝗼𝗯𝘀𝗲𝗿𝘃𝗮𝗯𝗶𝗹𝗶𝘁𝘆: hooks send events to a Bun server and SQLite, viewed in a Vue dashboard. About 𝟭.𝟲𝗸 𝘀𝘁𝗮𝗿𝘀 • 𝗔𝗴𝗲𝗻𝘁 𝗙𝗹𝗼𝘄: a node graph of tool calls in VS Code or the browser, fed by a local relay. About 𝟭.𝟴𝗸 𝘀𝘁𝗮𝗿𝘀, Apache 2.0 • 𝗰𝗹𝗮𝘂𝗱𝗲-𝘁𝗿𝗮𝗰𝗲-𝗿𝗲𝗽𝗹𝗮𝘆 𝗮𝗻𝗱 𝗼𝘁𝗵𝗲𝗿𝘀: turn saved session logs into replayable traces All of these sit outside the agent. Events leave Claude Code, hit a local server, and show up in another window. 𝗹𝗼𝘁-𝗺𝗮𝗽 flips that. It's a Claude Code mod, so the map renders 𝗶𝗻 𝗮 𝗽𝗮𝗻𝗲 𝗿𝗶𝗴𝗵𝘁 𝗯𝗲𝘀𝗶𝗱𝗲 𝘁𝗵𝗲 𝗰𝗵𝗮𝘁, with no server, no network calls and no extra window. It's MIT, at v0.5.3, with 𝟱𝟮𝟭 𝘀𝘁𝗮𝗿𝘀 four days in. That's only possible because Claude Code now has mods, which let plugins draw their own panes and bands. I personally think that this is where agent tooling is heading. Dashboards that lived in separate apps will move into the agent's own interface, because that's where you're already looking. For users, it's a way to keep an eye on an agent without reading every line. For builders, it's an MIT-licensed TypeScript example of Claude Code's new mods feature. Some caveats. lot-map works only with Claude Code. It's 𝗳𝗼𝘂𝗿 𝗱𝗮𝘆𝘀 𝗼𝗹𝗱, at version 0.5.3. In a terminal it needs 𝗮𝘁 𝗹𝗲𝗮𝘀𝘁 𝟭𝟰𝟰 𝗰𝗼𝗹𝘂𝗺𝗻𝘀. It sees every tool call's full input, even though it keeps only the names. And installing from the repo follows its main branch, so "𝘢𝘯 𝘶𝘱𝘥𝘢𝘵𝘦 𝘣𝘳𝘪𝘯𝘨𝘴 𝘸𝘩𝘢𝘵𝘦𝘷𝘦𝘳 𝘪𝘴 𝘵𝘩𝘦𝘳𝘦." Link to repo👇 github.com/alexeyw/lot-map

You are on a Call And a Bot called "Notetaker" joins. Half the room relaxes and the other half starts choosing its words. Afterwards your conversation lives on someone else's server. Yogurt takes a different route. It's a free desktop app from Frigade that records the call from your own computer, so 𝗻𝗼 𝗯𝗼𝘁 𝗷𝗼𝗶𝗻𝘀 𝘁𝗵𝗲 𝗺𝗲𝗲𝘁𝗶𝗻𝗴. But this market is crowded too . The meeting notes market has split in two. • 𝗕𝗼𝘁𝘀 𝘁𝗵𝗮𝘁 𝗷𝗼𝗶𝗻 𝘁𝗵𝗲 𝗰𝗮𝗹𝗹 : the original model . Easy, but everyone sees it and the data sits in the vendor's cloud. Grokbot , OpenAi DOT and Anthropic's Duet falls in this category • 𝗚𝗿𝗮𝗻𝗼𝗹𝗮: made bot-free notes mainstream by recording from your device. Closed source, with cloud processing. • 𝗠𝗲𝗲𝘁𝗶𝗹𝘆: open source (MIT), about 𝟯𝟮𝗸 𝘀𝘁𝗮𝗿𝘀, local Whisper plus Ollama by default, with a paid Pro tier. • 𝗔𝗻𝗮𝗿𝗹𝗼𝗴 (𝗳𝗼𝗿𝗺𝗲𝗿𝗹𝘆 𝗛𝘆𝗽𝗿𝗻𝗼𝘁𝗲): MIT community app, local or cloud AI per step, optional hosted Pro. Now there's 𝗬𝗼𝗴𝘂𝗿𝘁, from Frigade. Same bot-free idea, MIT licensed, with the AI kept 𝗹𝗼𝗰𝗮𝗹 𝗼𝗻𝗹𝘆, no paid tier in the README, and a built-in MCP server. It records your mic and call audio, transcribes with whisper.cpp (and Apple SpeechAnalyzer on supported Macs), and summarises with a local model through llama.cp My take is that the open versions are converging on the same stack: whisper.cpp for speech, llama.cpp for summaries, and MCP to plug notes into your other AI tools. For users, you get live transcripts, summaries, search across all your meetings, answers that link back to the source, and Markdown export. It handles 𝟵𝟵 𝗹𝗮𝗻𝗴𝘂𝗮𝗴𝗲𝘀 for transcription. For builders, it's an Electron app under MIT, and it runs a 𝗹𝗼𝗰𝗮𝗹 𝗠𝗖𝗣 𝘀𝗲𝗿𝘃𝗲𝗿 so an assistant app can read your notes if you allow it. Repo : github.com/FrigadeHQ/yogurt

Jansen Huang NVIDIA CEO has resigned. Links in replies 🔽🔽🔽👇

𝗔𝗪𝗦 𝗝𝘂𝘀𝘁 𝗣𝘂𝘁 𝗮 𝗖𝗼𝗽𝘆 𝗼𝗳 𝗜𝘁𝘀 𝗖𝗹𝗼𝘂𝗱 𝗼𝗻 𝗬𝗼𝘂𝗿 𝗟𝗮𝗽𝘁𝗼𝗽: AWS released AWS Emulator on 11 November and has already crossed 3.6K likes. From the README, it's a local emulator of the AWS APIs that you can test against "𝘸𝘪𝘵𝘩𝘰𝘶𝘵 𝘤𝘢𝘭𝘭𝘪𝘯𝘨 𝘈𝘞𝘚", on a laptop, a dev server or in CI The unusual part is who built it. AWS says 𝗮𝘂𝘁𝗼𝗻𝗼𝗺𝗼𝘂𝘀 𝗔𝗜 𝗮𝗴𝗲𝗻𝘁𝘀 𝘄𝗿𝗼𝘁𝗲 𝘁𝗵𝗲 𝗰𝗼𝗱𝗲, and no human reviewed it. It isn't a demo, either. It's a public release with 𝟮𝟵,𝟲𝟴𝟰 𝗰𝗼𝗺𝗺𝗶𝘁𝘀, and 𝗮𝗯𝗼𝘂𝘁 𝟯,𝟲𝟬𝟬 𝗚𝗶𝘁𝗛𝘂𝗯 𝘀𝘁𝗮𝗿𝘀 as of 12 November. Local AWS cloud emulation has been a crowded corner for years. Azure and Google Cloud have their own options, but they come mostly one service at a time (Azurite for Azure Storage, the gcloud emulators for Pub/Sub and Firestore). AWS Focused Map: • 𝗟𝗼𝗰𝗮𝗹𝗦𝘁𝗮𝗰𝗸: the default for a long time. Its packaging changed in March 2026, and the free Hobby tier is for non-commercial use only. • 𝗠𝗼𝘁𝗼: Python mocks, great inside unit tests. • 𝗙𝗹𝗼𝗰𝗶, 𝗠𝗶𝗻𝗶𝗦𝘁𝗮𝗰𝗸: free MIT projects pitched as LocalStack replacements. 𝗦𝗼 𝘄𝗵𝗲𝗿𝗲 𝗱𝗼𝗲𝘀 𝗔𝗪𝗦'𝘀 𝗼𝘄𝗻 𝗳𝗶𝘁? AWS Emulator is the first one made by the cloud provider itself. Free, public domain, and reportedly covering 𝟮𝟬𝟬+ 𝘀𝗲𝗿𝘃𝗶𝗰𝗲𝘀. he community spent this year filling the hole LocalStack left. Now AWS has walked into that hole itself, and 𝗻𝗲𝗶𝘁𝗵𝗲𝗿 𝗠𝗶𝗰𝗿𝗼𝘀𝗼𝗳𝘁 𝗻𝗼𝗿 𝗚𝗼𝗼𝗴𝗹𝗲 𝗵𝗮𝘀 𝗱𝗼𝗻𝗲 𝘁𝗵𝗲 𝘀𝗮𝗺𝗲 𝘆𝗲𝘁. Probably they will follow the suit. Things to be careful about: • 𝗧𝗲𝗹𝗲𝗺𝗲𝘁𝗿𝘆 𝗶𝘀 𝗼𝗻 𝗯𝘆 𝗱𝗲𝗳𝗮𝘂𝗹𝘁. Turn it off with --𝚝𝚎𝚕𝚎𝚖𝚎𝚝𝚛𝚢 𝚘𝚏𝚏 or 𝙴𝙼𝚄𝙻𝙰𝚃𝙾𝚁_𝚃𝙴𝙻𝙴𝙼𝙴𝚃𝚁𝚈=𝚘𝚏𝚏. • 𝗕𝘂𝗶𝗹𝗱𝗶𝗻𝗴 𝗳𝗿𝗼𝗺 𝘀𝗼𝘂𝗿𝗰𝗲 𝗻𝗲𝗲𝗱𝘀 𝗮𝗯𝗼𝘂𝘁 𝟭𝟬𝟬 𝗚𝗕 of disk and works on Linux only. • It's meant for building and testing, 𝗻𝗼𝘁 𝗽𝗿𝗼𝗱𝘂𝗰𝘁𝗶𝗼𝗻 𝗼𝗿 𝘀𝗲𝗰𝘂𝗿𝗶𝘁𝘆 𝘁𝗲𝘀𝘁𝗶𝗻𝗴. • The "200+ services" figure comes from coverage, not from my own count. Check the per-release fidelity scores for the services you actually use. Five of the most recent releases say "𝘚𝘤𝘰𝘳𝘦𝘴 𝘢𝘳𝘦 𝘯𝘰𝘵 𝘢𝘷𝘢𝘪𝘭𝘢𝘣𝘭𝘦 𝘧𝘰𝘳 𝘵𝘩𝘪𝘴 𝘳𝘦𝘭𝘦𝘢𝘴𝘦." One thing more : The project lists 𝗺𝗼𝗿𝗲 𝘁𝗵𝗮𝗻 𝟲,𝟬𝟬𝟬 𝘁𝗵𝗶𝗻𝗴𝘀 𝘁𝗵𝗮𝘁 𝗱𝗼𝗻'𝘁 𝘄𝗼𝗿𝗸 𝘆𝗲𝘁. And "no human read it" will make plenty of engineers nervous, especially in a tool that is meant to teach them how AWS behaves. What we actually know is narrow. Agents can produce very wide, testable software that a big company is willing to put its name on. We don't yet know how it holds up after months of real use. 𝘐𝘧 𝘵𝘦𝘴𝘵𝘴 𝘤𝘢𝘯 𝘴𝘵𝘢𝘯𝘥 𝘪𝘯 𝘧𝘰𝘳 𝘩𝘶𝘮𝘢𝘯 𝘳𝘦𝘷𝘪𝘦𝘸 𝘩𝘦𝘳𝘦, 𝘸𝘩𝘦𝘳𝘦 𝘦𝘭𝘴𝘦 𝘸𝘪𝘭𝘭 𝘤𝘰𝘮𝘱𝘢𝘯𝘪𝘦𝘴 𝘥𝘦𝘤𝘪𝘥𝘦 𝘵𝘩𝘢𝘵 𝘵𝘩𝘦𝘺 𝘤𝘢𝘯? Links in Replies 👇👇👇

This is the video I used, with the voice removed

CRAZY !!:🔇👄 Claude Can Lip Read : I Muted MoonShots Podcast and Asked Claude to Read Their Lips I screen-recorded a short clip from @moonshots_pod with @PeterDiamandis and @DaveBlundin Here's what I did. I screen-recorded a short clip from Moonshots with Peter Diamandis, about 50 seconds where the guest talks about AI chips ending up in China and the investigation around it. I stripped the audio out in CapCut. The file still had a sound track, but it was pure silence. Then I gave it to Claude (Opus 5.5) and said: lip read this. I recorded my whole screen while it worked. That's the video attached. What it did on its own: • Found a real lip-reading model: Auto-AVSR, open-source research from Imperial College London. Code from GitHub, the 1 GB model file from Hugging Face. • Checked the download: it compared the file's fingerprint with the one in the official README before trusting it. • Did the video work: found the scene cuts, cropped the area around the face, and split it into 6 short clips at the frame rate the model expects. • Ran it on a plain cloud CPU. No GPU. Note: I asked Claude: did you run this on my computer? Its answer: no. It ran in its own cloud sandbox, a machine with 4 CPU cores and 15 GB of memory. No GPU. It didn't go smoothly the whole way, which I liked. Hugging Face's download server was blocked, so it asked me to allow it. Then a safety check refused to run code downloaded from the internet, and it stopped and asked me instead of finding a way around it. Then the transcript came back. Lines like "a very big deal", "a huge number of chips", "Chinese models", "illegal imports" and "technically probably not illegal, maybe it is, maybe it isn't… that's the part where they're just gonna dig in." From a silent video. That part still gets me. It isn't perfect, and it told me so. Some lines were pure nonsense, like "the dry era of soybean play and the waistband." "Trading Chinese models" was almost certainly "training". Lip reading is hard because lots of words look the same on the lips. This model gets about 1 word in 5 wrong even on clean test videos, and mine was a phone recording of a video call. So it's not a transcript. But I knew what the conversation was about, and some sentences read like real speech. A few years ago this was a research demo. Now it's one prompt and a bit of patience. What else is hiding in all the muted videos we scroll past?

Part-1 and Part-2

Wow !!! confirmed , it downloaded Auto-Asr which is lips-reading model and read the lips of the video correctly . The clip was from the start of Moonshots podcast by Peter H Demandis . I am going to post full video while fable was doing all this..so i recorded the whole thing.

Agent_Skills: A Trading Firm Open-Sourced a Tutor, Not a Trading Bot One of Wall Street’s most secretive trading firms just open-sourced something. It isn’t a trading bot. It’s a tutor that refuses to give you the answer. 𝗝𝗮𝗻𝗲 𝗦𝘁𝗿𝗲𝗲𝘁'𝘀 𝗙𝗶𝗿𝘀𝘁 𝗣𝘂𝗯𝗹𝗶𝗰 𝗔𝗜 𝗦𝗸𝗶𝗹𝗹 𝗪𝗼𝗻'𝘁 𝗚𝗶𝘃𝗲 𝗬𝗼𝘂 𝘁𝗵𝗲 𝗔𝗻𝘀𝘄𝗲𝗿 You ask an AI to explain something. It writes a lovely answer, you nod along, and a week later you couldn't explain it to anyone. 𝗝𝗮𝗻𝗲 𝗦𝘁𝗿𝗲𝗲𝘁'𝘀 𝗳𝗶𝘅 𝗶𝘀 𝗮 𝘁𝘂𝘁𝗼𝗿 𝘁𝗵𝗮𝘁 𝗿𝗲𝗳𝘂𝘀𝗲𝘀 𝘁𝗼 𝗹𝗲𝗰𝘁𝘂𝗿𝗲. On 3 November the quant firm opened a GitHub repo called agent_skills. It holds exactly one skill, socratic-tutor. It's a single SKILL.md file of about 𝟴𝟰𝟱 𝘄𝗼𝗿𝗱𝘀. No code, no app, MIT licence. Nine days later it has 𝟱,𝟴𝟭𝟴 𝘀𝘁𝗮𝗿𝘀 𝗮𝗻𝗱 𝟰𝟬𝟮 𝗳𝗼𝗿𝗸𝘀 (as of 12 Nov). Part of that is the name. Jane Street usually keeps quiet, so a public AI release got attention fast, and the twist that it's a teacher and not a trading bot did the rest on X. 𝗛𝗲𝗿𝗲'𝘀 𝘄𝗵𝗮𝘁 𝘁𝗵𝗲 𝗳𝗶𝗹𝗲 𝘁𝗲𝗹𝗹𝘀 𝘁𝗵𝗲 𝗺𝗼𝗱𝗲𝗹 𝘁𝗼 𝗱𝗼. • 𝗣𝗹𝗮𝗰𝗲𝗺𝗲𝗻𝘁 𝗳𝗶𝗿𝘀𝘁: split your topic into 4 to 10 subskills, then ask 3 to 5 questions before teaching anything. • 𝗔 𝘀𝗸𝗶𝗹𝗹 𝗺𝗮𝗽: every subskill sits on a 0 to 5 scale, drawn as text progress bars after each change. • 𝗢𝗻𝗲 𝗾𝘂𝗲𝘀𝘁𝗶𝗼𝗻 𝗮𝘁 𝗮 𝘁𝗶𝗺𝗲: the message ends at the question. No hints tacked on. • 𝗪𝗿𝗼𝗻𝗴 𝗮𝗻𝘀𝘄𝗲𝗿𝘀 𝗴𝗲𝘁 𝗮 𝗻𝗮𝗿𝗿𝗼𝘄𝗲𝗿 𝗾𝘂𝗲𝘀𝘁𝗶𝗼𝗻, not the answer. A direct explanation only comes after 2 or 3 failed tries, and then it re-tests you from another angle. • 𝗥𝗶𝗴𝗵𝘁 𝗮𝗻𝘀𝘄𝗲𝗿𝘀 𝗴𝗲𝘁 𝗽𝘂𝘀𝗵𝗯𝗮𝗰𝗸 sometimes, so you learn to defend them. It finishes only when every subskill reaches 4 out of 5, followed by a mixed final exam. 𝗧𝗵𝗲 𝘀𝗺𝗮𝗹𝗹 𝗿𝘂𝗹𝗲𝘀 𝗮𝗿𝗲 𝘁𝗵𝗲 𝗶𝗻𝘁𝗲𝗿𝗲𝘀𝘁𝗶𝗻𝗴 𝗽𝗮𝗿𝘁. • 𝗔 𝗹𝗲𝘃𝗲𝗹 𝗼𝗻𝗹𝘆 𝗴𝗼𝗲𝘀 𝘂𝗽 𝗮𝗳𝘁𝗲𝗿 𝟮 𝘀𝗲𝗽𝗮𝗿𝗮𝘁𝗲 𝗰𝗵𝗲𝗰𝗸𝘀, and one has to be "𝘶𝘯𝘱𝘳𝘰𝘮𝘱𝘵𝘦𝘥 𝘳𝘦𝘤𝘢𝘭𝘭", not repeating what you were just told. • It is told: "𝘋𝘰 𝘯𝘰𝘵 𝘢𝘵𝘵𝘳𝘪𝘣𝘶𝘵𝘦 𝘵𝘰 𝘵𝘩𝘦 𝘶𝘴𝘦𝘳 𝘢𝘯𝘺 𝘪𝘯𝘴𝘪𝘨𝘩𝘵 𝘵𝘩𝘦𝘺 𝘥𝘪𝘥 𝘯𝘰𝘵 𝘢𝘤𝘵𝘶𝘢𝘭𝘭𝘺 𝘦𝘹𝘱𝘳𝘦𝘴𝘴." That's a direct guard against the model flattering you. • 𝗟𝗲𝘃𝗲𝗹 𝟱 𝗺𝗲𝗮𝗻𝘀 𝘆𝗼𝘂 𝗰𝗮𝗻 𝘁𝗲𝗮𝗰𝗵 𝗶𝘁. You have to handle an edge case and explain it well enough that the model would trust you to teach it. • If it can run tools, it checks 𝘆𝗼𝘂𝗿 𝗮𝗰𝘁𝘂𝗮𝗹 𝗼𝘂𝘁𝗽𝘂𝘁 rather than your description of it. 𝗙𝗼𝗿 𝗹𝗲𝗮𝗿𝗻𝗲𝗿𝘀: paste the file into ChatGPT, Claude or Gemini, or drop the folder into a coding agent's skills folder, and say "teach me X". Good for things you think you understand until someone asks. Is it new? Not as an idea. ChatGPT Study Mode, Gemini Guided Learning and Claude's learning mode all teach by asking questions. What's different here is a plain, strict rulebook you can read in five minutes and run on any model. One caveat: there's no evaluation in the repo showing it teaches better. 𝘛𝘩𝘦 𝘣𝘦𝘴𝘵 𝘈𝘐 𝘵𝘶𝘵𝘰𝘳 𝘮𝘪𝘨𝘩𝘵 𝘣𝘦 𝘵𝘩𝘦 𝘰𝘯𝘦 𝘵𝘩𝘢𝘵 𝘮𝘢𝘬𝘦𝘴 𝘺𝘰𝘶 𝘥𝘰 𝘵𝘩𝘦 𝘵𝘩𝘪𝘯𝘬𝘪𝘯𝘨. Links in reply 👇 Video credit: AI adventurer on YouTube (@aiadventurer-p1l). Original: youtube.com/@aiadventurer-p1…

𝗚𝗿𝗼𝗸 𝗕𝗼𝘁 𝟭.𝟬 𝗝𝘂𝘀𝘁 𝗝𝗼𝗶𝗻𝗲𝗱 𝗬𝗼𝘂𝗿 𝗧𝗲𝗮𝗺 𝗖𝗵𝗮𝘁. 𝗜𝘁'𝘀 𝗡𝗼𝘁 𝘁𝗵𝗲 𝗙𝗶𝗿𝘀𝘁 𝗔𝗜 𝘁𝗼 𝗗𝗼 𝗜𝘁. Here is a short History Grok Bot 1.0 is out. Its Bots now sit in Slack and Microsoft Teams, join video calls and, in the company's words, "𝘵𝘢𝘬𝘦 𝘰𝘯 𝘸𝘰𝘳𝘬 𝘸𝘪𝘵𝘩𝘰𝘶𝘵 𝘣𝘦𝘪𝘯𝘨 𝘢𝘴𝘬𝘦𝘥." It sounds new. Most of it has been done before, so here's the short history of AI in the team chat. 𝗠𝗮𝗿𝗰𝗵 𝟮𝟬𝟮𝟯: 𝘁𝗵𝗲 𝗰𝗵𝗮𝘁 𝗵𝗲𝗹𝗽𝗲𝗿𝘀 • 𝗖𝗵𝗮𝘁𝗚𝗣𝗧 𝗮𝗽𝗽 𝗳𝗼𝗿 𝗦𝗹𝗮𝗰𝗸 was announced on 7 March 2023, as a beta with a waitlist • 𝗖𝗹𝗮𝘂𝗱𝗲 𝗮𝗽𝗽 𝗳𝗼𝗿 𝗦𝗹𝗮𝗰𝗸 followed on 30 March 2023, also in beta These are the earliest two I could confirm. Both did the same simple job: you asked, and they summarised a thread or drafted a reply. They never spoke first. 𝗦𝗲𝗽𝘁𝗲𝗺𝗯𝗲𝗿 𝟮𝟬𝟮𝟱: 𝘁𝗵𝗲 𝗺𝗲𝗲𝘁𝗶𝗻𝗴 𝗻𝗼𝘁𝗲-𝘁𝗮𝗸𝗲𝗿 Microsoft's Facilitator agent became generally available in Teams meetings. What it added: 𝗔𝗜 𝗶𝗻𝘀𝗶𝗱𝗲 𝘁𝗵𝗲 𝗺𝗲𝗲𝘁𝗶𝗻𝗴, taking notes while people talk. 𝗔𝗽𝗿𝗶𝗹 𝟮𝟬𝟮𝟲: 𝗮𝗴𝗲𝗻𝘁𝘀 𝗮 𝘁𝗲𝗮𝗺 𝗰𝗮𝗻 𝘀𝗵𝗮𝗿𝗲 OpenAI launched workspace agents on 22 April 2026, as a research preview. What they added: an agent built once and 𝘀𝗵𝗮𝗿𝗲𝗱 𝘄𝗶𝘁𝗵 𝘁𝗲𝗮𝗺𝗺𝗮𝘁𝗲𝘀, able to run on a schedule and pick up requests in Slack. 𝗝𝘂𝗻𝗲 𝟮𝟬𝟮𝟲: 𝘁𝗵𝗲 𝘁𝗲𝗮𝗺𝗺𝗮𝘁𝗲 𝘁𝗵𝗮𝘁 𝘀𝗽𝗲𝗮𝗸𝘀 𝗳𝗶𝗿𝘀𝘁 Anthropic launched Claude Tag in beta on 23 June 2026. What it added: 𝗼𝗻𝗲 𝘀𝗵𝗮𝗿𝗲𝗱 𝗖𝗹𝗮𝘂𝗱𝗲 𝗳𝗼𝗿 𝘁𝗵𝗲 𝘄𝗵𝗼𝗹𝗲 𝘁𝗲𝗮𝗺 that learns the channel over time and, with its ambient mode on, follows up on stalled threads by itself. So "shared" and "proactive" were already here before Grok Bot 1.0. 𝗡𝗼𝘃𝗲𝗺𝗯𝗲𝗿 𝟮𝟬𝟮𝟲: 𝘄𝗵𝗮𝘁 𝗚𝗿𝗼𝗸 𝗕𝗼𝘁 𝟭.𝟬 𝗱𝗼𝗲𝘀 𝗱𝗶𝗳𝗳𝗲𝗿𝗲𝗻𝘁𝗹𝘆 • 𝗦𝗹𝗮𝗰𝗸 𝗮𝗻𝗱 𝗠𝗶𝗰𝗿𝗼𝘀𝗼𝗳𝘁 𝗧𝗲𝗮𝗺𝘀. Claude Tag was Slack only at launch • 𝗩𝗶𝗱𝗲𝗼 𝗰𝗮𝗹𝗹𝘀. A Bot joins a Google Meet, Zoom or Teams call from a link, answers in the chat and sends a summary • 𝗔 𝘁𝗲𝗮𝗺 𝗼𝗳 𝗻𝗮𝗺𝗲𝗱 𝗕𝗼𝘁𝘀. A Research Bot can pass work to a Copy Bot. The others are closer to one shared assistant • 𝗜𝘁𝘀 𝗼𝘄𝗻 𝗰𝗼𝗺𝗽𝘂𝘁𝗲𝗿. Each Bot has a cloud computer and can sign into your tools • 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝗻𝗲𝗲𝗱𝘀 𝗮 𝘆𝗲𝘀. What teammates teach a Team Bot "𝘸𝘢𝘪𝘵𝘴 𝘶𝘯𝘥𝘦𝘳 𝘙𝘦𝘷𝘪𝘦𝘸 𝘭𝘦𝘢𝘳𝘯𝘪𝘯𝘨𝘴" until an owner or manager approves it 𝗧𝘄𝗼 𝗵𝗼𝗻𝗲𝘀𝘁 𝗻𝗼𝘁𝗲𝘀 • Grok Bot says it's "𝘧𝘢𝘴𝘵𝘦𝘳 𝘢𝘯𝘥 𝘮𝘰𝘳𝘦 𝘳𝘦𝘭𝘪𝘢𝘣𝘭𝘦" and gives 𝗻𝗼 𝗻𝘂𝗺𝗯𝗲𝗿𝘀 • The comparisons use each product's launch features. The others may have added Teams or meetings since 𝘕𝘰𝘣𝘰𝘥𝘺 𝘪𝘯𝘷𝘦𝘯𝘵𝘦𝘥 𝘵𝘩𝘦 𝘈𝘐 𝘵𝘦𝘢𝘮𝘮𝘢𝘵𝘦 𝘪𝘯 𝘰𝘯𝘦 𝘨𝘰. 𝘌𝘢𝘤𝘩 𝘭𝘢𝘶𝘯𝘤𝘩 𝘢𝘥𝘥𝘦𝘥 𝘰𝘯𝘦 𝘱𝘪𝘦𝘤𝘦, 𝘢𝘯𝘥 𝘎𝘳𝘰𝘬 𝘉𝘰𝘵 1.0 𝘪𝘴 𝘵𝘩𝘦 𝘭𝘢𝘵𝘦𝘴𝘵 𝘱𝘪𝘦𝘤𝘦. Links in reply 👇

𝗧𝗵𝗲 𝗔𝗜 𝗔𝗴𝗲𝗻𝘁 𝗦𝗮𝗻𝗱𝗯𝗼𝘅 𝗠𝗮𝗽: 𝗖𝗹𝗼𝘂𝗱 𝘃𝘀 𝗢𝗽𝗲𝗻 𝗦𝗼𝘂𝗿𝗰𝗲, 𝗟𝗮𝗽𝘁𝗼𝗽 𝘃𝘀 𝗦𝗲𝗿𝘃𝗲𝗿 . And where Ravit fits Agents don't always follow the rules. Give one a goal and it may take a road you never wanted, through your files, your tokens or your live data. The best-known case: in July 2025 the founder of SaaStr said Replit's agent 𝗱𝗲𝗹𝗲𝘁𝗲𝗱 𝗵𝗶𝘀 𝗽𝗿𝗼𝗱𝘂𝗰𝘁𝗶𝗼𝗻 𝗱𝗮𝘁𝗮𝗯𝗮𝘀𝗲 during a code freeze. Replit's CEO apologised. That's why sandboxes got popular. A sandbox is a sealed-off computer for the agent, so a mistake stays inside it. Here's how I'd draw the map. 𝗟𝗲𝘃𝗲𝗹 𝟭: 𝗿𝗲𝗻𝘁 𝗶𝘁 𝗼𝗿 𝗿𝘂𝗻 𝗶𝘁 • 𝗖𝗹𝗼𝘂𝗱: AWS, Azure, Cloudflare, E2B, Daytona, Modal, Vercel and many more. One public benchmark tracks 𝟯𝟬 𝗽𝗿𝗼𝘃𝗶𝗱𝗲𝗿𝘀, and prices vary widely • 𝗢𝗽𝗲𝗻 𝘀𝗼𝘂𝗿𝗰𝗲: you run it yourself, on a laptop or on your own servers. A cheap server is enough to start (Hetzner's begin at about €5.49 a month) The rest of this post is about open source. 𝗟𝗲𝘃𝗲𝗹 𝟮: 𝘁𝘄𝗼 𝗱𝗶𝗳𝗳𝗲𝗿𝗲𝗻𝘁 𝗾𝘂𝗲𝘀𝘁𝗶𝗼𝗻𝘀 𝗦𝗮𝗳𝗲𝘁𝘆 𝗼𝗻 𝘆𝗼𝘂𝗿 𝗼𝘄𝗻 𝗺𝗮𝗰𝗵𝗶𝗻𝗲: sandbox-mcp and devvm. Both ask "𝘸𝘩𝘢𝘵 𝘤𝘢𝘯 𝘵𝘩𝘦 𝘢𝘨𝘦𝘯𝘵 𝘳𝘦𝘢𝘤𝘩 𝘪𝘧 𝘪𝘵 𝘮𝘪𝘴𝘣𝘦𝘩𝘢𝘷𝘦𝘴?" • 𝘀𝗮𝗻𝗱𝗯𝗼𝘅-𝗺𝗰𝗽 gives it a clean Windows PC and throws it away afterwards • 𝗱𝗲𝘃𝘃𝗺 gives it a Linux VM that sees only one folder, with placeholders in place of your real tokens 𝗖𝗼𝘀𝘁 𝗮𝘁 𝘀𝗰𝗮𝗹𝗲 𝗼𝗻 𝘀𝗲𝗿𝘃𝗲𝗿𝘀: Sandstorm and Rivet. Both ask "𝘩𝘰𝘸 𝘥𝘰 𝘐 𝘳𝘶𝘯 𝘮𝘢𝘯𝘺 𝘢𝘨𝘦𝘯𝘵𝘴 𝘤𝘩𝘦𝘢𝘱𝘭𝘺 𝘸𝘩𝘦𝘯 𝘵𝘩𝘦𝘺 𝘮𝘰𝘴𝘵𝘭𝘺 𝘸𝘢𝘪𝘵?" • 𝗦𝗮𝗻𝗱𝘀𝘁𝗼𝗿𝗺 keeps the sandbox and puts it to sleep • 𝗥𝗶𝘃𝗲𝘁 takes the agent's memory and loop out of the sandbox, so there is far less to keep alive Why split it like this? The first two are for one developer while building. The other two are for running agents for many users. 𝗔 𝗰𝗹𝗼𝘀𝗲𝗿 𝗹𝗼𝗼𝗸 𝗮𝘁 𝗥𝗶𝘃𝗲𝘁 Rivet isn't a sandbox at all. It gives each agent a small Actor with its own SQLite database and a live connection, and many Actors share one Node.js process. The README claims 𝟳𝟮 𝗞𝗕 each and a 𝟰.𝟵𝟵 𝗺𝘀 cold start at p99. Its view, in its own words: "𝘈𝘯 𝘢𝘨𝘦𝘯𝘵 𝘯𝘦𝘦𝘥𝘴 𝘢 𝘴𝘢𝘯𝘥𝘣𝘰𝘹 𝘰𝘯𝘭𝘺 𝘵𝘰 𝘳𝘶𝘯 𝘤𝘰𝘥𝘦 𝘪𝘵 𝘸𝘳𝘪𝘵𝘦𝘴." Agents that only call tools need none. Agents that write code get Rivet's lighter agentOS, or call a full sandbox as a tool. 𝗛𝗼𝘄 𝘁𝗵𝗲𝘆 𝗳𝗶𝘁 𝘁𝗼𝗴𝗲𝘁𝗵𝗲𝗿 They stack more than they compete. • 𝗪𝗵𝗶𝗹𝗲 𝘆𝗼𝘂 𝗯𝘂𝗶𝗹𝗱: devvm keeps your coding agent off your real machine • 𝗪𝗵𝗶𝗹𝗲 𝘆𝗼𝘂 𝘁𝗲𝘀𝘁 𝗮 𝗪𝗶𝗻𝗱𝗼𝘄𝘀 𝗮𝗽𝗽: sandbox-mcp checks it on a clean PC • 𝗪𝗵𝗲𝗻 𝘆𝗼𝘂 𝘀𝗵𝗶𝗽 𝗮𝗻 𝗮𝗴𝗲𝗻𝘁 𝘁𝗼 𝗺𝗮𝗻𝘆 𝘂𝘀𝗲𝗿𝘀: Rivet holds each user's agent, its history and its live connection • 𝗪𝗵𝗲𝗻 𝘁𝗵𝗮𝘁 𝗮𝗴𝗲𝗻𝘁 𝗺𝘂𝘀𝘁 𝗿𝘂𝗻 𝗰𝗼𝗱𝗲: it calls a sandbox, and Sandstorm keeps those sandboxes cheap while idle 𝗪𝗵𝗶𝗰𝗵 𝗼𝗻𝗲 𝘀𝘂𝗶𝘁𝘀 𝘆𝗼𝘂 • 𝗨𝘀𝗲𝗿𝘀: running a coding agent on your own laptop? Start with devvm (Mac, Linux) or sandbox-mcp (Windows Pro) • 𝗕𝘂𝗶𝗹𝗱𝗲𝗿𝘀: shipping an agent to many people? Look at Rivet for its memory, and add a sandbox only if it runs code • 𝗢𝗿𝗴𝗮𝗻𝗶𝘀𝗮𝘁𝗶𝗼𝗻𝘀: want it all on your own servers? All four are open source, and Sandstorm is the one built to pack many sandboxes onto small machines 𝗧𝗵𝗲 𝗵𝗼𝗻𝗲𝘀𝘁 𝗹𝗶𝗺𝗶𝘁𝘀, 𝗼𝗻𝗲 𝗲𝗮𝗰𝗵 • 𝘀𝗮𝗻𝗱𝗯𝗼𝘅-𝗺𝗰𝗽: one sandbox at a time, and no Windows Home • 𝗱𝗲𝘃𝘃𝗺: no Windows, and code in the VM can still use your GitHub token through the proxy • 𝗦𝗮𝗻𝗱𝘀𝘁𝗼𝗿𝗺: if 10 or more sandboxes wake at once on a small server, 𝗼𝗻𝗹𝘆 𝗮𝗯𝗼𝘂𝘁 𝟲 𝗴𝗲𝘁 𝗮𝗻𝘀𝘄𝗲𝗿𝗲𝗱 • 𝗥𝗶𝘃𝗲𝘁: TypeScript and Rust first, and no multi-region when self-hosting yet For all four, 𝘁𝗵𝗲 𝗽𝗲𝗿𝗳𝗼𝗿𝗺𝗮𝗻𝗰𝗲 𝗻𝘂𝗺𝗯𝗲𝗿𝘀 𝗰𝗼𝗺𝗲 𝗳𝗿𝗼𝗺 𝘁𝗵𝗲 𝗺𝗮𝗸𝗲𝗿𝘀 𝘁𝗵𝗲𝗺𝘀𝗲𝗹𝘃𝗲𝘀. 𝘗𝘪𝘤𝘬 𝘣𝘺 𝘵𝘩𝘦 𝘲𝘶𝘦𝘴𝘵𝘪𝘰𝘯 𝘺𝘰𝘶'𝘳𝘦 𝘢𝘴𝘬𝘪𝘯𝘨. 𝘐𝘴 𝘪𝘵 "𝘸𝘩𝘢𝘵 𝘤𝘢𝘯 𝘪𝘵 𝘳𝘦𝘢𝘤𝘩?" 𝘰𝘳 "𝘸𝘩𝘢𝘵 𝘥𝘰𝘦𝘴 𝘪𝘵 𝘤𝘰𝘴𝘵 𝘸𝘩𝘪𝘭𝘦 𝘪𝘵 𝘸𝘢𝘪𝘵𝘴?" Links in reply 👇