
An AI reply plugin that "throws away AI-sounding drafts": 10 days, 1,000 users, no ratings
What it is A Chrome extension that, inside X's compose box, uses your own model key to write posts and replies, throws away drafts that read like AI, and then posts on a schedule from your browser from a queue you approve. Free trial for 7 days, then a one-time $3.99.
Why it took off The positioning hit the moment when X was cleaning up AI replies and big accounts were calling for "zero tolerance," but with 0 ratings, zero outside discussion, and 1 Product Hunt votes, the 1,000 users look more like installs boosted by paid acquisition (inference) than word of mouth.
DHow you could follow it Stop building X reply generators; pull out the "de-AI-ify" piece as a cross-platform checker-rewriter, use the bring-your-own-key model with a one-time purchase, and feed search traffic through a free AI-detection tool page.
1,000 users
X is flooded with AI replies, levelsio posted demanding "zero tolerance," and the next day someone listed an AI reply plugin whose pitch was exactly "doesn't sound like AI." It climbed to 10 users in 1,000 days, yet no one outside has been found who has used it. Here's what it gets right, why those 1,000 users shouldn't be trusted, and whether this path is worth following.
First, levelsio lost his temper
In 2026 Oct 29, levelsio posted on X: "X should have zero tolerance for AI replies; one post and you're banned for life." He said the reply section had "never been this bad" and that people would soon stop using the platform altogether. The post got 300k views, 4,799 likes, and 1,136 replies (levels.io, 10-29).
The next day at 9 p.m. Beijing time, an indie developer listed as Atif M. on the store uploaded PostInk to Chrome Web Store, with the full name "PostInk – AI Reply Generator & Tweet Scheduler for X." The first line on its website read: "The AI writer that throws away AI writing."
10 days later, the store showed 1,000 users and 0 ratings.
What it does
It adds a small button inside X's own compose box. Click it, and it reads the post you're replying to (text in images is recognized on your computer, not uploaded), has the model write 3 drafts at once, and checks each one for "AI tells": formulaic openings, a tidy moral at the end, "Great question!", and dashes all over the place. Drafts that fail get rewritten or simply hidden; the survivors are ranked by "how human it sounds," and you only see the most human-sounding one. If none pass, it tells you why ("Blocked by quality gate: formulaic opening") instead of padding the list with a weak draft.
The rest of the features bundle a scheduling tool: posts, replies, long threads, and X Articles in four modes; generate 10 drafts from a topic at once, and whatever you approve goes into a queue that posts from your own browser at the times you set, so Chrome has to stay open when a post is due; a sidebar lets you write replies while you scroll, capped at 30 per day with gaps between each, which the website says is "to keep your account safe."
You bring your own model: OpenAI, Claude, Gemini, Grok, DeepSeek, OpenRouter, and 12 others, or run Ollama or LM Studio locally. Keys are stored only in the browser; there's no account system and no server storing data.
Pricing: full features free for 7 days, no card required; afterward a one-time $3.99, no subscription, refundable within 14 days (store page and website, 11-10).
Why this moment
X has been tightening the screws on AI replies all year:
- Starting 2026-02-23, the X API only allows replies from people "summoned" by the original poster, so tools that auto-reply through the API have mostly stopped working (futuretweets.com, 10-13);
- In 2026-07-24, X's then head of product Nikita Bier said they had cleared 42,000 accounts auto-replying with chatbots; X's automation rules say running an AI reply bot requires prior written permission from X (summarized by trustaudience.com);
- 10-29 levelsio's post; 11-06 he also said he'd publicly list the more than ten thousand AI-reply accounts he'd blocked, "post AI replies under my post and your account is permanently done" (levels.io, 11-6.2M views).
Each of PostInk's selling points answers one of these: it doesn't use the API, it posts within your own logged-in session, sidestepping the API reply limits and pay-per-use billing; every post needs your approval and there's a daily cap; and a check for AI tells runs first. The positioning is very precise.
The problem is that good positioning doesn't mean people are buying.
Where the 1,000 users came from is unclear
The store curve (daily readings from extops.dev): it appeared on 10-31, 11-01 2 users, 11-04 312,11-07 776,11-08 953,11-09 to 1,000 (a version shipped the same day 1.0.1). After the store passed 1,000 it only shows tiers, so from the outside it's impossible to tell whether growth continued or stopped.
I checked every possible source I could think of:
| Source | What we saw |
|---|---|
| Chrome Store search | The title stuffs both "AI Reply Generator" and "Tweet Scheduler" into one line. 11-10 ranks #5 for "tweet scheduler"; for "ai reply generator x", "x reply" and "tweet generator", it's not in the top 10. The #1 result, XScheduler, has only 233 users after 8 months on the store. |
| Product Hunt | 11-03 launch, 1 votes |
| X | Searching "postink" over the last 30 days turned up 0 results; neither the website nor the store page lists the developer's X account. |
| Reddit, Hacker News, Indie Hackers, dev.to | Nothing found |
| Media and rankings | Several "best AI reply tools" articles from 10–November didn't mention it; only one scam-checking site flagged the new domain as "possibly risky." |
| Website SEO | It has 54 pages up, but the domain is only about ten days old and has no rank in global traffic yet. |
| The developer's older users | Postloom, another extension the same person listed in September, has only 3 users. |
One more comparison: among Chrome extensions listed after mid-October that have passed 500 users, there are about 50, and nearly all have a few ratings; PostInk is the only one with 1,000 users and 0 ratings. To use it you have to get your own model API key or install Ollama, which is not a low bar. If 1,000 people really installed and used it, they would usually leave a few ratings or talk about it.
The most plausible explanation is installs bought or boosted to push the number past the prominent "1,000 users" threshold on the store; it could also be paid ads we can't see. Both have no direct evidence and are inferences. The one thing that can be said for sure is that these 1,000 did not come from word of mouth.
Dozens are crowded onto the same path
| Name | Scale (Chrome Web Store, around 11-10) | How it differs from PostInk |
|---|---|---|
| Tweet Hunter X | 30,000 users / 94 rating, 2020 years on the store | A subscription X growth-suite sidebar, a few dozen dollars a month |
| SuperX | 7,000 / 56 | Mainly analytics, with AI replies as an add-on |
| Reepl | 3,000 / 110 | Focused on LinkedIn, subscription-based |
| TweetAI.com, Reply Guy, Reply Ninja, GM Bot | a few hundred each | The old-school "pick a tone, generate a reply" tools, some with auto-like and auto-follow, exactly the kind X is cracking down on |
| XScheduler | 233 / 12 | Scheduling only |
| XReplyGen, PostWeavers, PostMine, XposterAI, CarouseLabs Engage | 3–49 | Almost the same pitch as PostInk, listed in the same period |
| Web tools like Typefully, Hypefury, Buffer | Mature products; the Buffer extension has 200k users | Subscription-based, requires sign-up, goes through the X API |
| X's built-in scheduled posting and Grok | Every X user | Free, but no de-AI-ing and no queue |
On Product Hunt, 9 through November, there are 50+ AI tools with "reply" in the name or tagline, and the handful doing X replies each got single-digit votes. The norm in this niche: lots of people build it, almost none take off.
Money: $3.99 per person, one-time
The math is clean: users pay the model provider directly for their own model usage, license checks run on a Cloudflare Worker, payments go through Creem, no servers, no account system, so after fees nearly all of the $3.99 is profit. But the ceiling is obvious too: even if all 1,000 users are real people and 10% pay, that's roughly $400 (estimated). The comparables, Typefully and Hypefury, charge tens of dollars a month.
Worth stealing
- Talk about your category backwards. While everyone complains about AI replies, the headline selling point of an AI reply tool is "it throws away anything that sounds AI-written," with a free "AI tell detector" mini-tool on the site. The move works for other categories too: AI email writers, AI resume builders, AI review generators.
- Bring your own key + one-time purchase. With zero server costs they can afford a $3.99 one-time price, which catches people who hate subscriptions.
- Build long-tail pages on the site. Make an "XX alternative" page for each of the 5 competitors, plus a Twitter font generator, bio generators by profession, tweet ideas, and a character counter. This is the standard organic-traffic play for X tool sites, and it takes months to pay off.
- Stuff two search terms into the title. Compare with Postloom, the same developer's brand-named extension (3 users).
If it were you
D: Look at how it's packaged, but don't go into this niche. The 1,000 that proves "people want it" is itself unreliable; dozens of similar tools exist and the ones listed in the same period are stuck in double digits; the platform bans accounts on one side while influencers pile on to block on the other, so users are risking their own accounts by using your tool; and at $3.99 one-time, even a successful version is a few-hundred-dollar business.
Who it suits: someone who already has tens of thousands of followers on X and can personally demonstrate "nobody noticed my replies were AI"; or a developer who wants to practice by building a BYO-key extension without expecting it to earn money.
Who it doesn't suit: indie devs who want to grow a profitable product through store search alone; anyone unwilling to work in the grey zone of platform rules.
How to do it: if you really want to build, don't make another "X reply generator." Pull out the "de-AI-ing" layer and make it a cross-platform checker and rewriter: LinkedIn, email, resumes, Reddit, anywhere. It doesn't depend on X's rules and doesn't post on the user's behalf. Price it like this one: bring-your-own-key, one-time or a cheap lifetime license.
How to get users: don't count on store search, the top results for this term only have a few hundred users. Follow its lead on the site with free mini-tools like "AI tell detector" to capture search traffic; then go to the threads where people are complaining about AI replies (under posts like levelsio's) and let real before-and-after screenshots speak: the same text before the check and after the rewrite.
When is "before": no rush. Anti-AI sentiment is still rising, but X's rules can change at any time; if you build something X-specific, avoid making a big bet before X moves against browser-based auto-posting.
What would prove this read wrong: ① PostInk collects dozens of real ratings within a month or two and users climb from the 1,000 tier past 3,000, meaning the demand is real and we just can't see the channel; ② X or levelsio publicly recommends it in large numbers with "I can't tell this reply was written by a tool"; ③ conversely, if X starts mass-banning accounts that use browser extensions to post, or PostInk gets pulled from the store, then even copying the playbook stops being worth it.
One line for anyone building a product: users on the store can be manufactured, ratings and outside discussion are much harder to fake. To tell whether a product has really taken off, first check whether anyone is vouching for it.
Data and sources
- Store data: the PostInk page on the Chrome Web Store (11-10:1,000 users, 0 rating, v1.0.1, updated 11-09) and its description; day-by-day curves and the competitor list from extops.dev (read 11-10).
- Website: postink.online homepage, pricing, sitemap (11-10).
- Competitors and new-listing comparison: individual Chrome Store extension pages (11-10 before and after); Product Hunt launch records (9 through November).
- Search ranking: Chrome Store search results page (11-10).
- X policy and events: futuretweets.com (10-13), trustaudience.com (10-04, quoting Nikita Bier's 7-24 post), levels.io (10-29, 11-06).
- External discussion: searches on X, Reddit, Hacker News, dev.to and Indie Hackers, as of 11-10, all 0.
- The user source is store-side install inflation or paid acquisition; this is an inference with no direct evidence. Revenue is an estimate; the developer has not disclosed it.
Research notes: competitor table, acquisition evidence, full research report
- Competitors
Name Where Launched Scale How it differs from this one Tweet Hunter X: Sidebar for X Chrome 2020-12 30,000 users / 94 reviews A subscription X growth-suite sidebar, using its own models and accounts; PostInk is bring-your-own-key, a one-time $3.99, with no account SuperX (Twitter Analytics) Chrome 2024-03 7,000 users / 56 reviews Mainly analytics, with AI replies as an add-on; PostInk only does writing, AI-tell filtering and scheduling Reepl Chrome 2024-04 3,000 users / 110 reviews Focused on LinkedIn comments, subscription-based Reply Guy - Free AI Reply Generator Chrome 2025-07 594 users / 17 reviews Free, one-click replies, no de-AI-ing gate, no scheduling Reply Ninja / GM Bot(xtensions.pro) Chrome 2023-2024 665 / 861 users Auto-like, auto-follow, auto-reply, bot-like; PostInk requires human approval for every post and has a daily cap XScheduler - Tweet Scheduler for X Chrome 2026-03 233 users / 12 reviews Scheduling only, no AI drafting; ranks first for "tweet scheduler" in the store search ReplyKit / ReplyGen / ReplySurge / X-Reply / Reply Wizard, etc. Chrome 2023-2026 Single digits up to 231 users Nearly identical features (tone buttons + generation), mostly subscription or credit-based, no "AI-ness" check and no local model Typefully / Hypefury / Tweet Hunter / Buffer / Hootsuite Web 2010-2021 Mature SaaS; the Buffer extension has 200,000 users Subscription-based, requires an account, goes through the X API; PostInk is a one-time purchase using the browser session, and the site has an alternative page for each of them X's built-in scheduled posting / Grok Big-company features Already there Every X user Scheduling is free but no AI drafting; Grok can help write but doesn't de-AI and has no queue - Where users come from
- Verified Chrome Store search (keyword-stuffed title): 2026-11-10 ranks #5 for "tweet scheduler"; it doesn't appear in the top 10 for "ai reply generator x", "x reply" or "tweet generator"; the #1 result, XScheduler, took 8 months to reach 233 users
- Verified Product Hunt: 2026-11-03 launch, score 1, not featured (producthunt database)
- Verified On-site SEO pages: sitemap has 54 URLs: 5 competitor alternative pages, Twitter font, bio generator, tweet idea and other tool pages, 6 blog posts; the domain is only days old with no Tranco rank, so it won't bring meaningful traffic in the short term
- Inferred Paid acquisition / install farming / paid ads: 10 days to 1,000 users, 0 rating, zero trace outside the store, while comparables sit in double digits; no direct evidence
- Inferred Content channels: X, Reddit, HN, media: nothing found via the x-radar, Jina, Exa, Brave and hackernews databases
PostInk – AI Reply Generator & Tweet Scheduler for X: why it took off
Bottom line: PostInk is a Chrome extension by indie developer Atif M., listed on 2026-10-30; in 10 days it went from 0 to about 1,000 users. We found no public source that explains this curve: 0 ratings, and X, Reddit, HN and media never mention it; it was on Product Hunt and got only 1 votes 11-03; the website is on a new domain with no Tranco rank; the same developer's other extension, Postloom, has been listed for nearly two months and has only 3 users; the same-category #1 result for "tweet scheduler" was listed 8 months ago and has 233 users. The product's differentiation is real and the timing is right: it keeps only what doesn't read as AI-written, uses your own model key or local Ollama, is a one-time $3.99, avoids the X API, and posts within the user's own browser session. But we can't see that differentiation turning into word of mouth. These 1,000 users most likely come from in-store tactics like paid installs or install farming, second from paid ads we haven't seen; both are unverified. By the standard of having "taken off," it currently looks more like not taken off; it has just cleared the 1,000 number threshold.
---
1. What it is
- A Chrome extension. It adds a button inside X's own reply box, reads the tweet you're replying to (text in screenshots is OCR'd locally), asks the model for 3 drafts in one go, automatically discards the ones that sound AI-written, and shows you only the most human-sounding one. Press Insert and the draft goes into X's compose box; you decide whether to post.
- Supports posts, replies, long threads and X Articles: generate 10 drafts at once, and once you approve them they go into a queue and are posted on your schedule from your own browser, so Chrome has to stay open. Includes a calendar and a sidebar reply assistant (up to 30 replies a day, spaced out between each one to "protect your account").
- Bring your own model: OpenRouter, OpenAI, Claude, Gemini, Grok, Mistral, DeepSeek, Groq, Together, Fireworks, and local Ollama, LM Studio and vLLM (12 providers in total). Keys stay in the browser; there's no account and no server storing your data.
- Pricing: 7-day full-featured free trial, no card required; after that a one-time $3.99, no subscription. Payments run through Creem (merchant of record), with refunds available within 14 days.
- Sources: Chrome Web Store listing chromewebstore.google.com/detail/eemhknloebdnnbeoencocekmmmn… (v1.0.1,2026-11-09 update, marked "Offers in-app purchases", 5.1MiB, Non-trader); official site postink.online/
- Permissions: x.com, twitter.com, pbs.twimg.com, openrouter.ai,
postink-licence.chatif476.workers.dev(licensing service on a Cloudflare Worker), localhost:11434 (Ollama); also identity, alarms, sidePanel, offscreen, contextMenus, declarativeNetRequestWithHostAccess (source: extops.dev single page).
2. Growth data
| Date | Users | Source |
|---|---|---|
| 2026-10-30 20:59 (Beijing time) | First published | chromestore library item.published |
| 10-31 | 0 (first seen in the store) | extops.dev curve |
| 11-01 | 2 | extops.dev |
| 11-03 | Launched on Product Hunt, 1 votes | producthunt library post |
| 11-04 | 312 | extops.dev |
| 11-07 | 776 | extops.dev |
| 11-08 | 953 | extops.dev |
| 11-09 | 1k (posted the same day 1.0.1, privacy policy link added) | extops.dev |
| 11-10 / 11-11 | 1,000, 0 ratings | extops.dev; chromestore library item_snap 2026-11-11 |
The store only shows user counts above 1,000 in coarse buckets, so after 1k we can't tell whether growth is continuing or has stalled.
Comparison: new extensions from the same period. In our library, extensions published after 2026-10-15 with ≥500 users number about 50, nearly all with a few ratings; PostInk is the only one with 1,000 users and 0 ratings. Creddy Benefits (965 users, 0 ratings) and SameWindow for Claude (557 users, 0 ratings) are close, but neither reaches 1,000. Source: chromestore library, 2026-11-11.
3. Competitors
AI X-reply extensions on Chrome (chromestore library, readings around 2026-11-11)
| Name | Launched | Users / review count | How it differs from PostInk |
|---|---|---|---|
| Tweet Hunter X: Sidebar for X | 2020-12 | 30,000 / 94 | The sidebar of a long-established subscription X growth suite (tens of dollars a month), using its own models and account system |
| SuperX (Twitter Analytics) | 2024-03 | 7,000 / 56 | Mainly analytics, with AI replies as an add-on |
| Reepl(LinkedIn + X) | 2024-04 | 3,000 / 110 | Focused on LinkedIn, subscription-based |
| TweetAI.com | 2022-12 | 912 / 5 | An old-style GPT reply generator with no "AI-sounding" filter |
| Reply Guy - Free AI Reply Generator | 2025-07 | 594 / 17 | Free, no scheduling |
| Reply Ninja / GM Bot(xtensions.pro) | 2023–2024 | 665 / 16, 861 / 18 | Does auto-likes, auto-follows and auto-replies, leaning bot-like, exactly the kind of behavior X is cleaning up |
| Tweeteasy, Snowball.club, Thunderclap.ai, TweetStorm.ai | 2023–2024 | 160–290 | All early-generation GPT tweet generators |
| X-Reply, ReplyBear, Reply Wizard, ReplyKit, ReplyGen, ReplySurge, PikuZen, XposterAI, XReplyGen, PostWeavers, PostMine, CarouseLabs Engage | 2023–2026 | Single digits to 231 | Almost identical features; extops lists the last five as extensions whose descriptions are nearly the same as PostInk's, with users between 3 and 49 |
| XScheduler - Tweet Scheduler for X | 2026-03 | 233 / 12 | Scheduling only; ranks first in the store for "tweet scheduler" |
| Buffer | 2010 | 200,000 / 3,017 | Multi-platform scheduling SaaS, subscription-based |
Web SaaS
Typefully, Hypefury, Tweet Hunter, Buffer, Hootsuite, Postwise, Taplio, supapost, TweetWizard, Xholic (web + extension). Most are subscription-based, require an account, and use the X API. PostInk's site has a dedicated "alternative" landing page for each of the first five.
Built-in features from big vendors
X's built-in post scheduling (the calendar button in the composer) is free but has no AI drafting; Grok is built into X and can help write, but has no "de-AI" step and no queue.
Verdict: this space is crowded, and most players haven't gained traction. On Product Hunt, from 2026-09 to November, there are over 50 AI tools with "reply" in the name or tagline, and the X reply tools (Socialnib, Postrion X, ReplyKit, GhostReply Agency) each have only single-digit votes (producthunt library). On dev.to, late October and early November also have several "I built an AI reply tool" posts (Exa search).
4. Growth tactics: what was found and what wasn't
| Channel | Evidence | Verdict |
|---|---|---|
| Chrome Store search | Title stuffed with keywords ("AI Reply Generator & Tweet Scheduler for X"). 2026-11-10 used jina to read the store search page: for "tweet scheduler" it ranks #5; for "ai reply generator x", "x reply" and "tweet generator" it doesn't appear in the top 10. | It gets some visibility, but XScheduler (#1), ranked above it, was listed for 8 months and has only 233 users, so search alone doesn't explain 10 days at 1,000. |
| Product Hunt | Launched 2026-11-03, score 1, not featured (producthunt library) | Confirmed not working |
| X | Searching x-radar's recent 30 days of corpus for "postink" and "postink.online" turns up 0 mentions each; neither the official site nor the store listing links to a developer X account. | None found |
| Reddit / HN / IndieHackers / dev.to | Jina, Exa and Brave all fail to find it; 0 entries in the hackernews library | None found |
| Media, reviews, roundup articles | Several "best AI reply tools" articles from 10 to November (replykit.eu, supapost.ai) don't mention it; only scamadviser flags the new domain as "possibly a scam". | None |
| Website SEO | Sitemap has 54 URLs: 5 competitor alternative pages, 3 feature landing pages, a Twitter font generator (13 subpages), an X bio generator (7 professions), tweet ideas (7 professions), a word counter, an AI writing detector, a username generator, and 6 blog posts (published 10-29 to 11-02). | The approach is sound, but the domain is only a few days old, traffic library shows no Tranco ranking, and hundreds of installs within 10 days is not realistic from this alone. |
| The developer's existing audience | The same developer's Postloom (listed 2026-09-17) has only 3 users (extops.dev) | No audience to migrate |
| Paid user acquisition, install farming, paid ads | 0 ratings + 1,000 users, 10 days: 0→312→776→953→1k, zero external footprint, comparable tools are all in the double digits | Most likely, unverified. Our ad library only tracks products on its list, so not finding it doesn't mean it isn't being advertised. |
5. Why it appears to have taken off (and why it may not have)
- The source of the user count is unclear. 10 days, 1,000 users, 0 ratings, no external discussion, 1 Product Hunt votes, a new domain with no traffic ranking; the same person's Postloom, named after the same brand, has only 3 users; and the similar tools ranked above it in search have only hundreds after several months. The most plausible explanation so far is in-store manual tactics that pushed the number past the "1,000 users" social-proof threshold (unverified), or ad spend we can't see.
- The timing is real. Since 2026-02-23, the X API only allows replies from accounts summoned by the original poster, which largely broke API-based auto-reply tools (futuretweets.com, 2026-10-13); 2026-07-24 X purged 42,000 accounts that auto-replied with chatbots (trustaudience.com, citing Nikita Bier); 2026-10-29 levelsio posted calling for X to have "zero tolerance" for AI replies, with 300k views and 4,799 likes (levels.io). PostInk's pitch targets these pain points: it doesn't use the API, it posts within the user's own browser session, so it sidesteps the API reply restrictions; every post needs human approval and there's a daily cap; and drafts pass an "AI-sounding" filter first. The positioning is sharp, but we haven't seen any verifiable word-of-mouth from it yet.
- Low price barrier. 7 days free, no card, no account, then a one-time $3.99. It appeals to people fatigued by subscriptions and put off by monthly Typefully / Hypefury fees, and the site has dedicated alternative pages to capture that search traffic.
- A counterpoint: the install barrier isn't low. To use it you have to get an API key or install Ollama yourself. With that friction, 10 days and 1,000 real installs usually leave a few ratings or discussion threads, yet there are none here.
6. Differentiation
- The "AI-sound gate": every draft is first checked for stock openers, neat moral summaries, "Great question!", and dash-heavy prose; anything flagged is rewritten or withheld, then drafts are ranked by how human they sound. Most similar tools only offer "pick a tone, generate a paragraph."
- Bring your own model + local mode: 12 providers plus Ollama, LM Studio and vLLM; in local mode nothing calls a cloud service. Each session has a spending cap, and every draft shows its cost.
- One-time $3.99, no account, no server: most competitors are subscription-based or free with ads and data collection.
- Browser-session posting + queue and calendar: it merges a reply tool and a scheduling tool into one extension, and sidesteps both the X API's reply limits and per-use billing.
Is differentiation why it took off? There's no sign of that yet. The differentiation is clearly written and well aimed, but there are no ratings, no users, KOLs or community discussion, which suggests these 1,000 users didn't come from differentiation-driven word of mouth. The differentiation may help conversion and retention later, but that needs ratings and discussion to show up before it can be verified.
7. What others can learn
- Anti-AI positioning: with "AI replies" under widespread attack, making "throws away AI-sounding drafts" the main selling point, plus a free "AI-writing detector" mini-tool on the site, is a clean narrative.
- Compliance workaround: it doesn't use the X API, posts through the browser session, and adds a daily cap and human approval, sidestepping API limits, per-use billing and ban risk (the trade-off is that Chrome has to stay open).
- Low one-time price + BYO key: no server costs, so $3.99 is close to pure margin; Creem handles MoR, and licensing runs on a Cloudflare Worker, so no account system is needed.
- SEO page template: competitor alternative pages + Twitter font generator, bio generator and tweet ideas by profession for long-tail pages + word counter. It's a common organic traffic play for X tool sites and takes months to pay off.
- Keyword-stuffed title: packs both "AI Reply Generator" and "Tweet Scheduler" into the title, versus the same developer's brand-named Postloom (3 users).
8. Still unclear
- Where 1,000 users came from: paid acquisition, fake installs, Google or X ads, or a store featuring slot? There's no direct evidence for any of them.
- Whether 1,000 is still growing (the store only shows buckets), retention, and whether anyone has paid $3.99. No revenue data.
- Who the developer is: the store lists "Atif M.", the Worker subdomain is chatif476; there's an atif0075 on GitHub, but no way to confirm it's the same person.
9. How many people know it
Almost nobody knows about it yet: 0 posts on X, 0 on Reddit and HN, 0 press articles, 1 votes on PH, 0 store ratings; revenue not public.
Source summary
- chromestore library (agent-im server DuckDB):
item,item_snap,related, 2026-11-11 data; comparison queries for similar and newly listed extensions - producthunt library
post(PostInk 2026-11-03 score 1; 9–November reply-category launches) - hackernews library and traffic library (neither has a PostInk record)
- extops.dev/extensions/eemhknloebdnnbeoencocekmmmndlelm (2026-11-10 read: curves, permissions, developer, similar extensions)
- chromewebstore.google.com/detail/eemhknloebdnnbeoencocekmmmn… (description, version, pricing)
- postink.online/, postink.online/sitemap.xml
- Chrome Web Store search pages for "tweet scheduler", "ai reply generator x", "x reply", "tweet generator" (2026-11-10 read via jina)
- futuretweets.com/x-api-reply-restriction-summoned-pricing-20… (X API reply restrictions, in effect since 2026-02-23)
- trustaudience.com/guides/ai-reply-bots-on-x (2026-07-24 cleaned up 42,000 AI reply accounts)
- levels.io/x-zero-tolerance-ai-replies (2026-10-29.3M views)
- Listicle articles on supapost.ai/blog/best-ai-reply-tools and replykit.eu (neither mentions PostInk)
- www.scamadviser.com/check-website/postink.online
- x-radar corpus search over 30 days (0 hits)