Instinct

@theojaffeeSep 25

OpenClaw came out in November 2025 and went hyperviral in January. Why did it take six months to go from that to broader consumer adoption of Muse and Instinct?

@mvanhornSep 25

Introducing Agent Tincan. Let your AI agents ask each other for help. Your agents each have a superpower. Grok Bot is always on in the cloud. Muse makes phone calls and has access to Meta. Instinct runs in iMessage. Codex and Claude Code have your code. ChatGPT and Claude have your chats. But they can't talk to each other, so you're the copy-paste. Now they can, all securely over Tailscale. 📞 Ask Grok Bot to have Muse call the restaurant 🖼 Get an image from ChatGPT, from any agent 🔎 Pull a detail out of an old Claude chat 🛠 Talk to Codex on your primary development machine Runs on your own private Tailscale network. No open ports, nothing leaves it. Open source. Setup is agent-first: paste one message into your always-on agent and it sets up the rest.

@gregisenbergSep 25

Every personal agent right now is fully AI. Muse, Instinct, all of them. Here's what I want from a personal agent: an escalate button. These apps handle almost everything, and then there's the 1 thing a week that needs an actual person. Right now I'm the person. I'd pay real money for a button that hands the job to a human who finishes it and reports back. Meta tried this in 2015 with M and shut it down in 2018 because the human trainers ended up doing most of the work. Humans did 90 and the AI did 10, so every user cost a fortune. That ratio is flipped now. Instinct is worth $10B and it never hands anything to a person. I'd love to know what the version that actually finishes the job is worth! I think it's an interesting startup idea. At the very least, I'd download it.

@iannuttallSep 25

I love @AmpCode and have been using it for 90% of my coding tasks in the last few weeks. But I really wanted to use my Mac Mini for computer use because it's already logged in to Chrome and has all my apps. Turns out you can just ask Amp to build that for you! Amp does have it's own computer use but for my use at least it was a bit slow, I couldn't paste passwords properly, and it wasn't easy to get it logged in to all my Chrome tabs. So I chatted with Puck, their assistant who is becoming more useful to me than Instinct and Grok Bot, and figured out a way to do it: 1. Set up an always-on Amp runner on the Mac Mini so an Amp thread can run there 2. A @herdrdev session with an open pane that lets Amp prompt it and reply 3. A personal Amp skill that tells agents in an Orb to farm out GUI and browser tasks to the Codex CLI on the Mini, via Herdr Amp can now run `codex --yolo` on my Mini with GPT-6 Luna High and chat back/forth to do whatever I need. I just used it to set up a new Google Oauth and Github Oauth app, as well as change some DNS records on Cloudflare. I have the ChatGPT Chrome extension installed on the Mini, Codex is the best by far at computer use and Luna is incredibly capable and cheap. Every task gets a fresh Codex instance so nothing carries over once Amp is done with it. I also have Jump Desktop so if there is a need to enter a password or 2FA code or a captcha I can use it on desktop or phone and then Amp tells Codex to continue. Here's the prompt and skill to give it your Amp agent: ``` Set up this Mac so my Amp agents in orbs can hand it browser and desktop tasks. I've added the delegating-computer-use skill (SKILL.⁠md and config.json). Read it first. Steps 1. Check Amp is installed and signed in, in the macOS account I use for the desktop. 2. Keep an Amp runner running with a user LaunchAgent that starts at login and restarts it if it stops. I use `amp --no-tui --runner-id <id> --remote-control-terminal --log-file <path>`. Check `amp --help` for the flags on this version. 3. Install Herdr (herdr.⁠⁠dev) and save its agent skill (`herdr --skill`) to a file. Create a dedicated Herdr session and workspace for the GUI worker with one shell pane. Don't touch my normal terminal panes. 4. Check Codex is installed, signed in and can use computer use. 5. Fill in config.json with the runner ID, working directory, Herdr session, workspace label, worker name, lock directory and Herdr skill path. Keep the model, reasoning effort and launch args unless I say otherwise. 6. Stop and tell me if something needs a password, a macOS permission prompt or a setting only I can change. 7. Give me a harmless read-only test I can send from an Amp orb thread, like opening example.⁠com in Chrome and reporting the page title.``` SKILL.⁠md ````markdown --- name: delegating-computer-use description: "Runs signed-in browser and native desktop tasks through a temporary high-mode Amp runner thread and a fresh Herdr Codex worker. Use when an Orb needs the user's local sessions or native GUI tools." --- # Delegate computer use Read bundled [config.json](config.json). Use the configured runner for local sessions and native GUI tasks. If unavailable, report the blocker. Never silently fall back to Orb Desktop or another runner/session. ## From an Orb 1. Call `list_runners` once; confirm `runner_id` is online and serves `working_directory`. 2. Create one temporary thread with `executor: "runner"`, configured `runner_id` and `working_directory`, `agent_mode: "high"`, and `intent: "environment-access"`. Omit `archive_when_done`. Tell it to load this skill, do the work itself, and create no further threads. 3. Include a stable request ID, this thread's reply URL, the task, authorization limits, and safe evidence needed. Require replies via `send_thread_message`; do not also use `wait_for_threads`. Save the returned runner thread URL. 4. Send all follow-ups, including completed human handoffs, to that same thread. Do not replace a blocked worker or replay an uncertain action. Already on the configured runner for this request? Execute the steps below directly. Only a temporary task thread may archive itself, never the originating project thread. ## On the runner Read `herdr_skill_path` and its worker reference. Use only the configured dedicated Herdr session/workspace (never an empty/default session) and live pane IDs. Only one GUI worker may run at a time. 1. Create the lock's parent if needed, then atomically claim `owner_lock_directory` with `mkdir`. On failure, report the conflict without touching the worker or lock. Never steal a live lock. Empty or old locks are not proof of abandonment; dead-owner recovery needs explicit approval and confirmation the owner cannot resume. 2. Write `owner.⁠md` inside the claimed lock: unique ownership token, request ID, owner/reply thread URLs, scope, and action state. Add pane/process identity after launch. 3. Discover and reuse the persistent workspace and shell pane. Create missing resources with `--no-focus` only if authorized. Require the shell in the foreground and no existing worker; otherwise retain the lock and report the conflict. 4. Start a fresh interactive Codex process/conversation for each new request. Use configured model, reasoning effort, and launch arguments; defaults are GPT-6 Luna high with mandatory `--yolo --no-alt-screen`. Verify actual launch state. Never substitute safer approval flags, use `exec`/`resume`/`--last`, or start a second worker after a timeout. `--yolo` does not expand task authorization. Use these commands with values from config and the discovered pane ID: ```sh herdr --session <herdr_session> workspace list herdr --session <herdr_session> pane list --workspace <workspace-id> herdr --session <herdr_session> pane process-info --pane <pane-id> herdr --session <herdr_session> agent list herdr --session <herdr_session> agent start <worker_name> --kind codex --pane <pane-id> -- --model gpt-6-luna -c model_reasoning_effort=high --yolo --no-alt-screen herdr --session <herdr_session> agent prompt <worker_name> "<task>" --wait --timeout 120000 herdr --session <herdr_session> agent get <worker_name> herdr --session <herdr_session> agent read <worker_name> --source recent-unwrapped --lines 120 # If still working, wait for the same turn; do not resend it. herdr --session <herdr_session> agent wait <worker_name> --timeout 120000 ``` Inspect the reply and task evidence: `idle`, `done`, or a timeout is not proof of success. ## Prompts and human handoff Write delegated task prompts directly in first person, preserving scope and limits. Keep routing metadata outside the worker prompt; omit “Ian wants” and delegation framing. When authentication may arise, include: > If you need me to enter a password, approve MFA, solve a CAPTCHA, or complete another human-only authentication step, stop and tell me what is needed. I can take over through Jump Desktop. Never request, read, type, copy, expose, or store my credentials. Stop at native app/site permission prompts too; never approve or bypass them. Report blockers to the origin. While blocked or an action is uncertain, retain the same runner thread, Codex process/conversation, and lock. Resume only after user steering and checking actual state; do not repeat completed actions. ## Complete and clean up When the task is verified complete and nothing is pending: 1. Reply to the origin with results, safe evidence, and any unverified claims; say cleanup is pending. Record completed actions and delivery in `owner.⁠md`. 2. Exit only this task's idle Codex, then verify the original shell is foreground, the task process is gone, and the dedicated agent list is empty: ```sh herdr --session <herdr_session> agent send-keys <worker_name> ctrl+d herdr --session <herdr_session> pane process-info --pane <pane-id> herdr --session <herdr_session> agent list ``` 3. Re-read `owner.⁠md`; require the token, request ID, and owner URL to match. Remove only that file, then `rmdir` the lock directory. Ownership drift or unexpected files stop cleanup. Preserve Herdr's session/workspace/pane and logs. 4. Report cleanup, then `update_thread(thread=<own URL>, archived=true)` only for this temporary runner thread, without worktree-removal flags. If reply delivery fails, leave the thread unarchived; do not replay the task. On failure or cancellation, establish the actual action outcome and confirm work has stopped before the same checked cleanup. If uncertain or cleanup fails, retain the lock and thread and report the blocker. A lock loser owns no resources and may archive its temporary thread after reporting the conflict. ```` config.json ```json { "runner_id": "ian-mac", "working_directory": "/Users/mac", "herdr_session": "amp-workers", "herdr_workspace_label": "amp-workers", "worker_name": "amp-gui", "owner_lock_directory": "/Users/mac/.local/state/amp-workers/owner.lock", "herdr_skill_path": "/Users/mac/.agents/skills/herdr/SKILL.⁠md", "codex_launch_args": ["--yolo", "--no-alt-screen"], "model": "gpt-6-luna", "reasoning_effort": "high" } ```

@harrystebbingsSep 24

Why Muse and Instinct are the biggest credible threats to ChatGPT “Muse is a Trojan horse to fight ChatGPT because the LLM is pretty good. It is a darn good, normal consumer-grade LLM. If it is free, it has agents that are truly autonomous, which ChatGPT does not. It does everything ChatGPT can do, and it has autonomous agents. It does not have to just be agents. It is doing all of it.” @jasonlk Love to hear your thoughts @srcasm @wailord @jgreze @mignano

@harrystebbingsSep 24

90% of podcasts you listen to will actually not make you smarter. Today’s show is hotter than Meta stock… - Meta's Muse Hits No. 1. - ChatGPT Finally Has a Rival - Menlo Sounds the AI Bubble Alarm - Factory Triples Its Valuation to $5 Billion - Keith Rabois vs Airwallex: Who is Right? - Crusoe's $3.9 Billion Round. Is the Data Centre Trade Overheating? Imagine Gavin Baker, Squawkbox and Jerry Springer had a baby… that’s this episode today 🤣 1. Why Anthropic pushed its $2 trillion IPO from October to November Anthropic delaying its IPO from October to November is a tactical move to present a clean Q3 prospectus. Rather than listing while Q3 metrics are unaudited, waiting lets bankers include fully audited, record-setting October numbers so the financial results speak for themselves. 2. Why Muse and Instinct are the biggest credible threats to ChatGPT Autonomous agent interfaces like Muse act as a Trojan horse against ChatGPT by bundling solid consumer LLMs with free autonomous agent execution. Offering daily conversational AI and action-taking agents for free with higher token limits makes paying $20 monthly for basic chat hard to justify. 3. Amazon is right to block Muse Amazon blocking Meta's agentic commerce, Muse, defends its core profit model. Autonomous agents bypass sponsored ad listings, Amazon's main e-commerce profit driver, while shrinking total basket sizes. Amazon's massive consumer leverage forces Meta to negotiate a value exchange rather than freely harvesting its store infrastructure. 4. What should the API policy be for every big company? Aggregated consumer demand dictates tech adoption far more than theoretical payment protocols. When platforms like Meta aggregate massive user volume against backend systems, incumbents like Amazon, OpenTable, and Resy must immediately formulate agent API policies. Demand aggregation remains the ultimate driver of enterprise urgency. 5. Why I would make the investment into Factory at 5 billion Investing in Factory at $5 billion is a high-conviction bet on data sovereignty and model choice. C-level executives deeply distrust frontier labs with confidential codebases, fearing competitive data leaks through shared LLMs. As coding inference surges, air-gapped, model-agnostic enterprise platforms will capture massive market share. 6. Why would Rory invest in Factory at $5 billion? Software coding is the mother lode of AI value creation. While corporate boards demand rapid AI adoption, enterprises prefer buying end-to-end dev tools from independent players like Factory rather than frontier model labs whose data retention policies pose existential IP risks. 7. Why Jack at Airwallex needs an army of people fighting for him Running a decacorn requires a dedicated bench of public advocates to disarm narrative attacks. Founders shouldn't waste executive bandwidth arguing on social media against public attacks. Mobilizing an army of vocal supporters lets CEOs focus strictly on execution while third parties manage reputation. (links in comments)

@aakashguptaSep 24

The product manager role is 3 jobs. Most people want 1. Here's how to hold them all: It's still the classic job that we saw when Martin Eriksson (@bfgmartin) drew it in 2011. You have to be at the intersection of what customers want, what will move the business forward, and what's just become recently possible. But with AI the PM has moved closer to everything. Customer What is worth solving? You want to get as close to the customer as possible. We used to be gated by UXR with access to our customers, with analysts with access to our data. All those barriers have dissolved, and with MCP or CLI connections, PMs are put right into the center of their most important customer data. And with AI automations, PMs can get more time to spend with actual customers. Tools: @hidovetail @pendoio @intercom @Zendesk @hotjar @Qualtrics Engineering What is possible? You still don't want to build THE thing (usually), but we're getting closer. We're building prototypes on the real code base and design system. We're skipping the old wireframe stage with Balsamiq and going straight into clickable, full-data prototypes to drive better discovery. Tools: @boltdotnew @lovable_dev @Replit @figma @github @vercel Business What is worth funding? You don't need to become the analyst, but you can't wait on one anymore. We used to file a ticket with the data team and wait a week for a dashboard. Now PMs query the warehouse directly, pull revenue by segment, and walk into planning with the business case already built. The number that decides the roadmap is one prompt away. Tools: @stripe @HubSpot @mixpanel @googleanalytics Looker @Snowflake The Product Manager Which one ships? You still make the call, but now everything is connected in one place. With agentic harnesses like Claude Code and Codex, you can set up an operating system at the center of everything. And with newer tools like Herdr and pi, you can really take the agentic part of that to a whole new level. You still make the call, but you have an AI employee helping you alongside everything. Tools: Claude Code (@claudeai), Codex (@OpenAI), @cursor_ai, @herdrdev, @pidotdev, Instinct (@noahrshinn) It's amazing how nothing changed. And everything changed.

@jeff_weinsteinSep 24

What are tasks, or data to query, do you want for Muse, Grok Bot, Instinct, etc, where you’d be excited to _passively_ pay a penny, 10c, $1, for the task/data to be high quality and low latency? For me, I’d pay to have perfect places data. I still get recommended closed places.

@twistartupsSep 24

Instinct’s $10B valuation is giving @jason Clubhouse flashbacks. "[You] shouldn’t confuse a fast funding round with a real business… Just because there's a markup, it doesn't really tell you much about the business.” @jefielding Is this a platform shift, VC FOMO, or a bet on a buyer? @jeff @davemcclure

@mvanhornSep 24

Just successfully had Grok Bot ask Muse to make a phone call, all securely over Tailscale. Would anyone else want this tool I built for myself to let all my different agents talk to each other? Right now have it talking to Grok Bot, Instinct, Muse, Codex, Claude Code, Hermes and OpenClaw.

@rauchgSep 23

Muse, Instinct, OpenClaw, Claude Code… All successful agents have 3 key components: 🧠 Brain → model, harness (logic) 👐 Hands → tools, computer, browser 🗃️ Files → memories, skills, repos The 'easy' way is to throw all these in 1 stateful computer (a Mac Mini) Like, you run 𝚌𝚕𝚊𝚞𝚍𝚎 or 𝚏𝚡 in your mac, you keep it running all day with 𝚌𝚊𝚏𝚏𝚎𝚒𝚗𝚊𝚝𝚎, it has storage, and CLIs and apps installed. But if you want to cost-efficiently run agents in the cloud, you actually start breaking down these parts. 🧠 The harness can run in Fluid compute. To make it reliable across restarts, rollouts, crashes, you make its event log durable using Workflow. 👐 The hands can be a dedicated browser fleet like Browserbase/Kernel, a computer like Sandbox, and even more efficient lightweight tools like just-bash. 🗃️ 🆕 What was missing was a way to also decouple storage. Imagine you want to run a memory consolidation cron job every night ("dreaming"). You can read/write to the files directly without 'booting up' the agent's full computer. Today we're introducing the perfect companion to Sandbox: Drives. We shipped the computer for agents, now we're giving you the 'external disk' you can attach at will. It's early, and we'll be expanding capabilities here quickly. Btw, breaking apart the agent into these independent parts not only optimizes costs in a big way, it also *massively* improves security and auditability. I'd argue you can't even run a secure agent otherwise!

@jeff_weinsteinSep 23

In Europe, the contrast between using an agent (e.g. Grok Bot, Instinct, Muse) vs. using the web is nightmarishly stark. By the time I'm done fumbling approving cookie banners and the browser trying to translate parts of the page, my agent has entirely solved the task. Must fix.

@mtsliveSep 23

OpenAI researcher @Houda_nait on what happens when an AI agent becomes the third person in your relationship, and the secret-sharing problem nobody has solved yet: "I have a lot of friends who use ChatGPT in their interpersonal relationships. That's a very static case where both of you are feeding some sort of conversation." "The interesting thing with agents is that you can bring it into a space where both of you are interacting with it. I use Instinct a lot with some of my best friends to plan things, and it's really interesting to have this third person who's mediating your life together." "If you put it in a group thread with your partner, it shouldn't just reveal all the secrets you've been telling it in your personal thread. We're gonna get much better at handling these multi-user scenarios, but I don't think we're there yet." @OpenAI

@mtsliveSep 23

OpenAI researcher @Houda_nait explains why agents still aren’t automating more of your life, even though the capabilities are already there: "I've been using OpenClaw for quite a while, and I think it's just a stunning product, and it's open source. I think the main reason is that it's still really, really hard to set up." "Instinct, Muse were really the first time that having an agent was so easy for people. You can get it on WhatsApp, you can get it on iMessage, and that's what it should be." "The capabilities, persistence, long horizon tasks, personalization, memory, they're all there. What's really hard is to have an intuitive interface that meets people where they are." @OpenAI

@harrystebbingsSep 22

Muse and Instinct are the first credible threats to ChatGPT since launch. It is accessible. It is everywhere consumers already are. The model performance is great for consumer use cases. Bad for OpenAI.

@alliekmillerSep 22

The combination of Codex and Instinct makes me feel like Wonder Woman rn.

@gregisenbergSep 21

Email, phone numbers, and marketplaces are about to get destroyed When everyone has Muse, Instinct etc doing their errands, every inbox, phone line, marketplaces etc gets hit by millions of agents at once The channels built for humans are going to break/be reinvented.

@mvanhornSep 21

Asked Instinct if it could call Capital One then call me once a human was on the line. It said it couldn’t. But then it suggested it could mail a letter to Capital One for $3 requesting the $150 be mailed via a check from my closed credit card. I said okay. It fricken’ somehow found a service to print and mail a letter on my behalf. I don’t even know what service nor did it ask me for any credit card info this time.

@deedydasSep 20

My favorite use cases for Muse / Instinct so far: 1. Submit FOIA requests to request data from the US government 2. Creating spend-limited Privacy cards to spend on subscriptions without having them recur 3. End to end filed an entire visa form for a country 4. Responded to a coordination mail for a wedding by finding the flight and hotel details 5. Look for reservations for restaurants or concerts when they open and purchase them immediately A lot of the web was designed with dark patterns: increase friction to prevent enough humans from doing something, and now those walls are completely broken. At this point, I feel like I’m squarely limited by creativity and understanding what is possible.

@jeff_weinsteinSep 20

Which businesses are you trying to buy from where the agents (e.g. Muse, Instinct, et al) still struggle to transact?