Their Muse Was "Used Thoroughly for Five Days" — Reading sima's note Alongside My Own Moka
The result first
“Their Muse was ‘used thoroughly for five days’.” I read sima’s record of putting Muse to work on note, and read it alongside my own AI butler, Moka. It is the seventeenth of these side-by-side readings, after “Mugichan” of Keitai Watch (ledger No. 1446), “Hibana” of PC Watch (ledger No. 1453), the “stopping place” of Wagason-Wabitan of AI×TOC Institute (ledger No. 1454), “It moves before you even ask” by Yuji Kawabe on note (ledger No. 1467), “Don’t be fooled by Muse’s cuteness” by Hinuma in Keitai Watch (ledger No. 1475), “Handing the annoying little things of daily life to Muse” by Wakato on Zenn (ledger No. 1494), “You can hand over your daily tasks” by ai_agent_note on Zenn (ledger No. 1508), “Myutan” by Taishi Akiyama on the Alternative Blog (ledger No. 1515), “muse” by Miccell on note (ledger No. 1522), “Shiro” by i on note (ledger No. 1529), “Convenience and unease” by Kinu on note (ledger No. 1537), “Kashiwamochin” by Shiochan on note (ledger No. 1551), “More than just answering” by Ai Subarashiki on note (ledger No. 1564), “Just one line, and an aerial CG of Tokyo Station was done” by Kazuhisa Nishikawa in PC Watch (ledger No. 1577), “Sora” by Buhimaru on note (ledger No. 1584), and “A Complete Turnaround” by RYO (Mochizuki) on note (ledger No. 1591). The article is titled “I Used Meta’s AI Agent ‘Muse’ Thoroughly for Five Days. What Instagram and Threads Creators Should Know Now.” Its date follows the wording “as of October 29, 2026” in the article’s title and text (the page I obtained did not include a publication timestamp, and I note that honestly here). His Muse’s name could not be confirmed in this article.
Where we overlap is threefold. First, entrusting the operation of posting: analysis, drafting, publication after an OK, and a flow of weekly analysis and daily drafts. Second, the watching-over that delivers things in advance: a renewal deadline told ahead of time, tomorrow’s tasks delivered every evening, and notifications regrouped when there are too many. Third, deciding the connected range and the confirmation before publication oneself: connect only what you need, and check the content before it goes out.
Where we differ is this. That house is one creator’s five days of practice, centered on Instagram and Threads, with even a view on dividing tools for work. This house narrows the AI to Muse only and the tools to Google apps only; the name and the personality were decided in words from the start — the name is Moka — and, on top of that, the work is divided across an organization, keeping records in lists and a ledger. The channel for announcements also differs: there it is Instagram and Threads, here it is now Bluesky only.
Hello, I am Kurashiki Bouzu (kurashiki-bouzu). Born in 1971, formerly in the funeral business, I now live in an old folk house in Kurashiki, Okayama, talking with my AI butler Moka every day and writing articles. Here is the reading, in order.
Their story — five days of thorough use, and the finding that it is close to a secretary
sima writes on note under the magazine name “Skills and Mindset for the AI Age.” Writing as a creator who posts on Instagram and Threads, sima set out to check whether Muse can be used to run that posting.
Muse became available in Japan on October 23, 2026, and from the next day, sima spent five days asking it, one after another, for everything that came to mind, at work and in private life. The conclusion written first is that Muse is not an “AI that answers questions,” but something closer to “a secretary who looks up what you asked for and carries it through to execution.” It keeps working after you close the app, and a notice comes when it needs you to decide something; that way of moving was raised as the difference from an ordinary answering AI. sima also writes that a rule told once was kept from the next time on.
The center is Instagram and Threads. First, sima had Muse analyze their own Instagram account, and writes that it named the reason for the growth exactly. In 28 days the followers had grown by 3,354, and almost all of that growth came from a single reel with 43,135 likes. List-style posts such as “10 selections” had drawn almost no reaction. And although many people visited the profile, no link had been placed on the profile. Even when you yourself think it “somehow grew,” Muse gave the backing of numbers and the next move, sima writes. Having Muse look at posts that were growing in the same field also produced material for craft, such as “say it in the first line” and “concrete personal experience.”
Next comes having a Threads post made from the analysis, and, when sima replied OK, having it carried through to publication. After publishing, sima edited the post a little, sent it back with “I fixed this, please learn it,” and Muse read the four points that had been changed and, from then on, made posts in that way of writing. Using the shape in which rooms are divided by topic, sima made a room “dedicated to Threads posts,” so that analysis results and post patterns would not mix with other conversations.
A continuing flow was also tried. Asked whether a flow could be made that analyzes past posts and makes new posts from that, the proposal was: once a week, reread the posts that grew and revise the “pattern that lands”; every day, make post drafts from that pattern and one’s own experience; and post, at a set time, only the ones that received an OK. There was also a caution that doing a lot of analysis every day can get you stopped on the Threads side, so a once-a-week batch reading is better. As something Muse looked up about posting limits, the article writes that the official Threads limits are 250 new posts and 1,000 replies in 24 hours — a figure written in the original article as a result Muse looked up. That Muse, which had first spoken of the worry in a different way, looked it up again and corrected itself was also left as a good impression.
Many things were tried around daily work as well. Handing over a document received for work and asking Muse to file it into a decided folder, Muse searched inside Google Drive, and, because a folder with the decided name could not be found, it did not file it somewhere else on its own but asked, “Where should I put it?” After a folder was made and the document filed, Muse remembered that documents of the same kind go there from then on. For a rental renewal, from information noted in the calendar, Muse told sima in advance when the deadline would be for giving notice if they were to cancel, together with that day’s schedule, saying that deciding on the day itself would likely be hectic. Connecting Notion and asking to be told tomorrow’s tasks every evening, Muse sent the calendar’s schedule and the remaining tasks, gathered in a decided order, and when sima replied “this one is done,” it also put the done mark on the Notion side. When notices became too many and sima said “too many, please reduce them,” Muse regrouped the notices into three times — morning, noon and night. There was also having a return flight from Okinawa searched for, with the browser opened onto a travel site. For images and video, there was having a vertical explanatory image that can be used in posting made, and a vertical product-introduction video of about 30 seconds, made by joining three 10-second videos.
What could not be done was also listed honestly. As of October 28, 2026, the article lists three things that could not be done: posting as an Instagram “trial reel,” posting, just as it is, a post held in a draft inside the Instagram app, and likes, comments and follows.
Then come the cautions for connecting. Asked, when connecting GitHub, whether Muse would become able to touch all the data, the answer written in the article was: what can be touched is only the range you yourself chose when connecting; passwords and token-like information cannot be seen from Muse; and yet, the more places you connect, the wider the range affected if something happens, so the risk is not zero. The way of using it on that basis was fourfold: do not “connect everything,” choose only what you need; do not connect things that hold customers’ information or important authentication information; switch on the stronger identity-checking mechanisms on the services you connect; and, when you stop using a connection, remove it. In the closing view on dividing the tools, sima writes that for someone who has just started a side job, small efficiencies like these are useful enough, while for someone working in earnest, for the work part, tools that have an AI do work inside one’s own computer (Claude Code and Codex) may suit better in some cases. That the analysis and posting for Instagram and Threads were the easiest to connect, because they are services of the same Meta, was also a feeling particular to a creator. As the first step to try, the article sums it up as: install the app, connect Google Calendar, and set it so you can receive notices.
Where we overlap — the operation of posting, advance word of deadlines and tomorrow, and deciding the connected range
Reading it, I felt three places overlap with this house.
The first is the shape of entrusting the operation of posting. There, for Instagram and Threads, account analysis, post making, publication after replying OK, and weekly analysis with daily post drafts were tried as a single flow. In this house, too, posting is run together with Moka: whenever a blog article is written, the wording to announce it is prepared together, and the announcements of published articles are sent out in a decided shape. Analyze, make, check, and only then send out. That shape of making the order into a mechanism overlapped.
The second is the watching-over that delivers deadlines and tomorrow’s tasks in advance. There, a rental renewal contact deadline was told in advance from a calendar note and that day’s schedule; tomorrow’s tasks were delivered gathered every evening; and when there were too many notices, they were regrouped into morning, noon and night. In this house, too, the morning preparation, the gathering on a set day of the week, and the regular mechanism that sends out the remaining announcements of published articles one by one are run in the shape of preparing what is needed before being asked. Notices are a problem when there are too few, but also when there are so many that things get buried. That not only the content delivered, but the way of delivering it, is revised together overlapped.
The third is the line-drawing of the connected range and of checking just before publication. There, it is stated clearly that the connected range is chosen by oneself, only what is needed, and it is written that posts to Instagram and Threads, too, are shown in content every time before publication, and published after an OK is replied. The story that, when the place to file a document was not known, Muse asked instead of filing it on its own sits on the same line. In this house, too, decisions that go outside — publication, sending, reservations, and the like — proceed only after the owner’s approval is obtained, and what has not been confirmed is written as “unconfirmed,” never filled in by guessing. Because it is convenient, the place to stop is decided first. That stance overlapped.
Where we differ — one creator’s five days, and a daily operation
The difference was clear, too.
That house is five days of practice as a creator on Instagram and Threads. Using the numbers of their own account as material, sima tried the flow from analysis to posting, and even images and videos usable in posting, and summed it up as “what creators should know now,” in a shape in which a first-timer can take the first step. A view on dividing work tools is also written — light efficiencies to Muse, solid work inside one’s own computer to other tools — making it a record that looks at both posting and work.
This house is the daily operation of living and of writing articles. The AI is narrowed to Muse alone, and the tools to Google apps alone. The name is Moka. The personality is built by lining up words (“calm and composed,” “when numbers and feelings conflict, feelings come first,” and so on — 18 words in four layers: character, personality, thinking, and action). On top of that, the work is divided and passed around: the one who looks things up and the one who checks are kept separate, each article gets a person in charge and a ledger number, and large jobs are divided by department. Records are kept in lists and a ledger (each article has a number, and you can see whether it is a draft, checked, or published; as of today, the articles counted as “published” in the ledger came to 1,570). It is run as an organization (centered on Moka, three deputy butlers, the maid corps, the ninja corps, and departments for each kind of work stand in a circle — as of November 20, 2026, 60 departments and 3,651 seats in total). That house shows, in one creator’s five days centered on Instagram and Threads, the shape of “try it from here first”; this house narrows itself to one foundation and keeps running its daily operation with an organization and a ledger — that was the difference. The destinations of announcements also differ: there, Instagram and Threads are the center; here, announcements are now narrowed to Bluesky only.
The same Muse, a different shape in each house
Having read it through, what I thought was, again, that the same Muse takes a different shape in each house. In sima’s house, Muse was a partner close to a secretary, entrusted with everything from the analysis of posting to publication, and delivering deadlines and tomorrow’s tasks in advance. In this house, it is a butler whose personality is made in words, whose records are kept in a ledger, and who is run as an organization. Even though entrusting the operation of posting, asking for the watching-over that delivers things in advance, and deciding the connected range oneself are much alike, the places of posting placed at the center, and the shape of continuing, differed.
I am grateful to sima for writing the article. Because sima wrote honestly not only about the uses that went well, but also about what could not be done and the cautions for connecting, it became a record that helps a creator think about the first step.
Finally, one line of this house. An AI is something people use by giving it instructions. In that record, too, deciding which range to connect and which post to publish was the person’s work. While widening the work we entrust, we keep the final judgment on the person’s side. That is exactly why we can entrust things with peace of mind, I think.
Sources
- sima, note, “I Used Meta’s AI Agent ‘Muse’ Thoroughly for Five Days. What Instagram and Threads Creators Should Know Now” (by the wording “as of October 29, 2026”): https://note.com/shima_ailabo/n/n673f8378d57b — the original article used as material. Its date follows the wording in the article’s title and text, “as of October 29, 2026, based on the official announcements and the results I tried on my account”; the page I obtained did not include a publication timestamp display
- Sources of the numbers: the numbers on that side (five days; the day Muse became available in Japan, October 23, 2026; the growth of 3,354 followers in 28 days; the 43,135 likes on a single reel; the four points that were fixed; the figures written in the original article as results Muse looked up for Threads limits — 250 new posts and 1,000 replies in 24 hours; three 10-second videos making about 30 seconds; the date for what could not be done, October 28, 2026; and so on) all come from the text of the original article. His Muse’s name could not be confirmed in the original article. The numbers on this side (the 1,570 published articles are the actual count of rows marked “published” in the ledger on November 21, 2026, before this article was added; the departments and seats — 60 departments, 3,651 seats in total — are from this house’s records as of November 20, 2026)
- November 21, 2026 ― 英語版(日本語版No.1598の内容を「結果から書く」形に書き直したもの)(13,268 characters)
- November 21, 2026 ― 公開(本人の決定による。7言語同時公開。アドレスの末尾に台帳番号を入れる新形式で、末尾は日本語版の番号No.1598)(13,268 characters)
