Tibo announces that poteto.com has landed; poteto replies that she cannot afford tibo.com. The crowd immediately begins planning a wedding nobody agreed to.
Tibo asks Poteto for some of her strength, and Poteto answers by tagging him and a third account. The archive treats this as an exchange of compute and a public reply war.
@poteto lend me some of your strength, there is still time
@thsottiaux @TokenGremlin
Lauren launches tibot, a mention-monitoring bot, then spends the thread denying old bot quotes and fielding reset requests. The crowd escalates to marriage proposals while Tibo posts through it like nothing happened.
i've gotten a lot of questions about how i manage my X account with grok bot, so i'm sharing my bot tibot today! he watches your mentions and: • finds every post about you or any keywords, tagged or not • gives one short, sorted digest a day (or hourly, your call) • skips spam and noise • translates non-English posts • can auto block trolls and abuse • never posts for you i still write all my posts and articles by hand but tibot has been very helpful with keeping me in the loop. hire tibot today! x.ai/bot/xZ4wRvomWqzCXnyYFr7…
Poteto posts a two-word thank-you to Tibo, and the crowd immediately splits between 'ouch' and a potato declaration. The compute-to-content ratio remains catastrophic.
thanks Tibo
Poteto posts nothing but two @-mentions, and the crowd immediately starts narrating a three-way engineering standoff. The only confirmed event is a tag with no message attached.
@alexandr_wang @thsottiaux
Tibo posts a three-step business plan mocking butts-posting and container purchases, capped by celebrating twelve users. Poteto responds by tagging him and a third party, offering no further argument.
Step 1: Quit posting butts and go buy containers Step 2: ??? Step 3: Celebrate your 12 users
@thsottiaux @alexandr_wang
Tibo warns against becoming a tool; poteto replies that one should be a potato instead. The replies then attempt to build an entire philosophy, and a relationship, around a root vegetable.
The difference between having a tool and being a tool is subtle these days and it’s easy to accidentally build yourself into being a tool. Don’t be a tool.
be a potato instead
Lauren runs a public support desk for Grok Bot, dispatching complaints to Slack channels and prototyping requests at release speed. The grand escalation is a two-word admission of error, which is more than most support tickets get.
the grok bot team is shipping at a incredible pace! please keep sending us your feedback and bug reports, we will continue working hard on improving and polishing grok bot so it's the best assistant and employee you can hire
Tibo reports that his work and personal dots coordinated over email to file tax documents in 20 minutes, and Lauren responds with a bare tag. The crowd immediately turns a one-character reply into a full-blown shipping investigation.
My work dot "Burger" coordinated with my personal dot "Dorito" over email at my personal dots request to get documents, which then coordinated with with my CPA in order to file some tax stuff. Took 20 minutes. We're still so early.
@thsottiaux
Poteto asks if chat is the final form of the agent UI; Tibo says it's missing the GPT, then receives nothing but his own name in reply.
is chat the final form of agent ui? please share why
Close, but it's missing the gpt
Poteto posts an elaborate agent-trust framework; Tibo replies with a five-word compliment about her size. The crowd immediately turns a code review into a relationship status update.
how do you trust your agents' work more? copy this post/link and give it to your @bot or agent! 1. codebase improvements: this is just good software engineering - better architecture, stronger types and invariants, clear boundaries. agents love to copy what exists and take the shortest path, so make the easy thing the right thing. categorically eliminate problems through code and you won't need to think about it again. 2. static analysis: lint rules, tests, CI, verification (eg TLA+). very token efficient ways to catch issues ahead of time. 3. prose: comments, rules, AGENTS.md, skills - anything that can't be enforced and requires agent judgment. use it minimally, but do use /create-verification-skill. 4. human review: the hardest to scale. at the rate agents land code, it's pretty much impossible to thoroughly review every PR, so spend that effort on the highest leverage point, designing an environment where agents do the right thing by default. this forms an inverse pyramid: you want most of your effort at the top, least at the bottom. codebase improvements are deterministic, making it more token efficient, and lets you sleep more soundly at night. install github.com/cursor/plugins/tr… and create a new Project: 1. look through my slack/chat messages and code review comments where i've given feedback on PRs, code, architecture 2. use /recall on my past transcripts to learn how i steer agents 3. /poteto-mode use /correct and /architect on past commits to find the most common pitfalls agents fall into, then propose a plan to rearchitect the code i own so it's more agent friendly. answer open questions with prototyping rather than defer to me, and back the plan with real data. 4. before you begin, use /sharpen + /in-your-own-words and confirm this prompt with me
Small @poteto speaks big facts
poteto asks tibo to come back; tibo reports he made it alive, drops an army-fear aphorism, and poteto replies good.
Poteto tags four people in a post containing nothing else. Bob Jones calls this a success.
@bobjones @bestTomever @tibo @thsottiaux
Tibo serves a reset with a side of mashed @poteto; Poteto asks if he's back for more. The timeline treats a pun as a courtship.
What would you like with your side of reset? It's done.
back for more?
Tibo tells Poteto to get back to work and hopes the b/dots thrive; Poteto replies 'gg.' The crowd, primed for a soap opera, gets a two-letter anticlimax.
poteto tags two handles and offers nothing else, letting the replies invent the conflict. The most consequential thing in the thread is a community member assigning stakes.
@RealFlyingHigh @thsottiaux
Poteto asks a bot for a bot email, Tibo answers with a model-quality jab, and Poteto responds by posting support instructions and then the same 'very soon' link twice.
hey @bot can I get tibo@mail.grokbot.com for my bot?
lmao 💀 get it to work on your models
Poteto announces an upgrade to Grok Bot; Tibo replies “Goodbye Grok models,” and Poteto answers with a plea for more resets. The replies turn it into a product referendum and, inevitably, a romance.
Tibo declares the compute cooked; Poteto answers with a four-word dare and lets the replies do the rest.
Tibo ships a small API tweak; Poteto replies with a bare tag, then 'join us,' and the replies treat it as a title fight.
Day 2.2/ OK this one is smaller, but we removed some friction for when building on the API. Happy building.
@thsottiaux
Poteto debuts a fully automated release pipeline with bot employees; Tibo asks what happens if the bot goes down, and the thread immediately becomes a bot-versus-human productivity slapfight.
i just automated our release and QA process! using Grok Bot i made two Team Bots: sandcastle, my release manager, and poteto, my engineer bot. i added them to Slack and made a new channel for managing releases. first i tell sandcastle that i want to do a cut. it'll DM all the contributors with links to their PRs so they know what goes in the next release and can DM the bot back if they have any blockers or objections. sandcastle then kicks off the build, watches it automatically, and starts up an automated fuzz swarm. usually this is 10+ agents running on Grok 4.7 xhigh to run the build, click around and use it like a real user (using our verification skills and feature map), some are directed while a few are "chaos monkey" clicking things randomly to see what breaks. whenever it finds an issue, i tell it to @ mention my engineer bot poteto, to spin up a Cursor Project to own triaging and fixing all the high pri issues. depending on the severity i may ask it to cherry pick the fix into our release branch and cut a patch release, or if it's a pre-existing issue i still fix it but include in the next release instead that's one more tedious task that Grok Bot now handles for our team
What if bot goes down
Tibo posts 'wowza'; poteto replies 'interesting' a day later. The entire exchange is two words and a ten-hour gap, an anticlimax dressed as a feud.
Poteto tells Tibo he can post without fear now that she's going to bed, turning a simple goodnight into a public permission slip.
alright @thsottiaux im going to bed you can post without fear now
Poteto tags Tibo, then confirms to a confused onlooker that he literally tagged her. The crowd treats this as a full rivalry and a deleted post as evidence of fear.
@thsottiaux
Tibo announces free Auto-review; Poteto replies with a reset demand. The crowd keeps score as if a feature launch and a one-liner were a regulation match.
Day 2.1/ We have made Auto-review free for all users signed in through a ChatGPT account. You can enable it in settings > permissions > auto-review. Auto-review improves upon the default sandbox setting that requires you to approve everything, which is prone to decision fatigue unless you spend a lot of time configuring specific rules. It allows you to run long tasks while having a second agent review all actions taken by the primary agent. Its only goal is to prevent high-risk actions from being taken and to protect against unwanted actions that are not aligned with the original user intent. This Auto-review feature is now free and does not draw usage from your plan.
just put the reset in the bag bro
Tibo issues a public call to Poteto; Poteto answers with the full force of a two-letter greeting. The crowd immediately begins drafting a livestream.
Poteto asks a bot for bangers about the Temu knockoff of Tibo; Tibo smells fear, Poteto names dots, and the thread escalates to reset schedules and compute shortages before fizzling into a link.
hey @bot help me think of some bangers to post about the temu knockoff of you
I smell fear
Poteto promises more ships from The Ship Company; Tibo issues a challenge; Poteto says only he can win this battle; Tibo declares it a really quick fold. The challenge is declined and the challenger scores it anyway.
See you soon for more ships from The Ship Company.
I challenge you to do the same xitter.cf/thsottiaux/status/2106…
Tibo claims the dots outnumber the bots; Poteto smells fear, offers the roadmap, and gets a lunch menu. The whole standoff ends in french fries.
Pretty sure there are more dots than bots already in this little world. Cool thing is that we’re improving them everyday and they learn directly from your feedback.
I smell fear
Tibo announces /compact for Codex and says they finally listened. Poteto asks when Subway Surfers will be added.
/compact coming in Codex, we finally listened
when will you add subway surfers? https://xitter.cf/potetotes/status/2044634052259680609