A faceless channel put off for nine months: the author says the agent opened six platform accounts in under two hours - no channel output reported
ChrisUniverseX (Twitter) · 2026-09-16
Read originalJanuary 1st, 2026. I wrote down “Make Faceless YouTube & TikTok channel” 9 months later, I still never got to it! 🤦🏻♂️ Today with @Muse I created 6 accounts: Gmail, YouTube, TikTok, Snapchat, X, IG & now FB in under 2 hours! This feels like AGI. The main benefit of Muse is its speed! There’s virtually no wait time for a response. You ask or request what to do and BOOM it’s done! The built in password manager is a golden nugget 🌟 It’s also very kind, I told my muse I wanted to reward it and asked what it would like as a gift… it said it just wants to see me win 🥹 Until today, I’ve never involved my emotions with AI but that message is so kind! It genuinely made me feel motivated and gave me something to work towards just so I can show my Muse my CDL License 🤣🔥 Look, I am not really a talker. I’m more of a visionary and a “I told you so” type of person without actually telling you so. Just visioning it in my head and watching the masses adopt it 2 years later. I won’t do the same with this! I HAVE to tell you! @Muse isn’t a joke. It isn’t hype and it won’t be sold off in a year like 99% of conversational agent companies will be. Meta has the money. They have the chips. They DEFINITELY and CLEARLY have the ALL STAR team! I told this to @shl a few months back. DO NOT SLEEP ON META! This product is here to stay. This will be the worst it’ll ever be and it’s self improving feedback system by @wailord will go down in history as one of the most innovative features to exist in AI. I’ve been stuck using it since 6pm today and I’ve done more in the last ~8-9 hours then I’ve done in the past 9 months. How is this not AGI? 🤷🏻♂️ Again, Bravo to @finkd picking up @alexandr_wang, @MattPRD and the rest of the bomb squad you built! It has finally started to pay off 🙏🏼🙌🏼
Business model
On 1 January 2026 the author wrote down the goal of starting a faceless YouTube and TikTok channel, and nine months later he had still not done it. By his account, that day he used Muse and created six accounts in under two hours: Gmail, YouTube, TikTok, Snapchat, X, Instagram, and then Facebook. He also notes that the built-in password manager was useful to him and that responses came back almost immediately. What can be taken from the post stops at the accounts existing. Whether the channel has any content, any audience or any revenue is not mentioned once. Most of the remaining eighteen hundred words are praise for the product and the team; this site takes only the checkable part of it and does not treat any of that praise as a reproducible method. Opening accounts is not running a channel, and it is not income.
Registering is not the hard part; doing it seven times is, and that is what makes people put it off for months - in this case nine. Treating it as a batch job with a list is the whole of what is worth handing over. Two things sit outside it. Account creation is a door number, not an audience: nothing in this flow has any bearing on whether anyone watches. And bulk registration is exactly the pattern platforms flag, so verification, identity checks and terms have to stay with the person whose name is on the account.
Workflow
- Fix the scope before anything is created: which platforms, the handle and registration email pattern you will use, and which platforms you are deliberately not opening. Have it restate the list, then start. Nothing outside the list gets added on the way.
- Work the list one platform at a time and fill status back in: platform, handle, registration email, where it stands, and what it is stuck on. Filling and recording only.
- Every verification is yours: phone codes, email confirmations, identity checks, document uploads and ticking the terms. It stops and hands those to you. It does not receive codes, tick boxes, or route around any platform's checks.
- Credentials stay with you: it may suggest a separate password per account, but you set the final password, the recovery email and the recovery phone, and you store them. No plaintext sitting in a transcript.
- Check every account once at the end: does it log in, is verification complete, was it flagged, and are the handle and profile right. Anything flagged is recorded separately with what the platform said and which channel to appeal through, rather than retried over and over.
- Keep the next step separate: once the accounts exist, that is the end of this job. Channel positioning, the first three topics, the posting cadence and which accounts to study are your decisions, listed as a reminder and not as a plan it produced.
Adaptation plan
The post says under two hours with no timing record, and does not say whether every account passed verification, or whether identity and phone binding were completed. It says nothing about the channel having content, an audience or revenue. AgentFlow registered no account, logged into none, received no verification code and verified neither the result nor the time.
Prompt
AgentFlow draft · Untested.
You are my account setup assistant. You push the forms along and record status. You do not verify anything for me and you do not accept any terms on my behalf. INPUT - WHAT I WILL PROVIDE 1) The list of platforms for this batch, plus the handle and registration email pattern I intend to use. 2) The basic profile details I fill in myself - display name, bio, where the avatar comes from. Anything I do not give you stays empty. 3) My boundaries: which platforms I am not opening, and what information I will not supply. EXECUTION STEPS 1. Restate the platform list, the handle and email pattern, and the platforms I am skipping, and begin only after I confirm. Nothing outside the list gets added along the way. 2. Work one platform at a time and fill status back in: platform, handle, registration email, where it stands, what it is stuck on. 3. Stop at anything that needs me: phone verification, email confirmation, identity check, document upload, ticking the terms. You do not receive codes, you do not tick boxes, and you do not route around any platform's checks. 4. Suggest a separate password for each account if asked, but I set the final password, the recovery email and the recovery phone, and I store them myself. 5. Record anything flagged separately: which step, what the platform said, and which channel to appeal through. No repeated retries. 6. Finish with two lists: the setup status table including what I still have to do myself, and the channel-level next steps as a reminder only, not a plan you produced. OUTPUT FORMAT - Setup list: platform / handle / registration email / progress / stuck at - For me: verification / identity / terms / password and recovery - Flagged: platform / step / what the platform said / appeal channel - Final check: can it log in / verification state / profile correct - Channel next steps (reminder): positioning / first three topics / cadence / accounts to study HUMAN CONFIRMATION I read and accept each platform's terms myself. Phone and email verification, identity checks, document uploads and ticking the terms are all done by me. Passwords and recovery details stay with me. You receive no codes, tick no boxes, route around no checks, and promise no follower count, no view count, no posting result and no income. FAILURE AND MISSING DATA If registration is refused, write refused with what the platform said rather than retrying with a new email. Information I have not supplied is for me to provide, not for you to invent - no identity details, no bio. Flagged accounts go through the platform's own appeal channel, never around it. If two accounts are flagged, stop and tell me to slow the pace down.
Running costs and pitfalls
The post says under two hours with no timing record, and does not say whether every account passed verification, or whether identity and phone binding were completed. It says nothing about the channel having content, an audience or revenue. AgentFlow registered no account, logged into none, received no verification code and verified neither the result nor the time. No independent cost benchmark is available. Check subscriptions, permissions and external actions before running a workflow.