Paid Founding 100 · Six agents · Approval-first
One crew covers SEO & GEO, scheduling, listening, creative, and ads — a complete marketing team from CLI, MCP, or the dashboard. Every draft lands in one queue; by default, nothing publishes until you say yes.
paid Founding 100 access when your invite activates · Cancel anytime
Works where you already work
How it looks
Everything your agents drafted, scheduled and shipped. Nothing posts until you approve it.
We rewrote auth this week. No new screens, no launch, nothing you would notice.
That was the point. Sign-in was our most-reported bug, and the fix was boring: one session model instead of three.
If you are putting off a boring rewrite, here is how we scoped ours.
We tried three tools before building ours, so here is the honest version.
If you have under 500 users, a plain-text sequence from your own inbox beats any template. Disclosure: I build a product in this space.
Your pricing page answers six questions in plain text, but none of them are marked up.
Here is the JSON-LD we would add, and why answer engines quote pages that have it.
Illustrative preview of the real page layout. No live data.
Tasks never publish content.
Nothing here yet. Save an ad from the Ad Library or track a competitor to get started.
No campaigns yet.
Finds opportunities, trends, and conversations while optimizing your brand for AI search
Illustrative preview of the real agent card. Nothing runs here.
Tracks metrics and recommends optimizations
Illustrative preview of the real agent card. Nothing runs here.
Posts content to social platforms at optimal times
Illustrative preview of the real agent card. Nothing runs here.
Creates platform-specific marketing content in your voice
Runs when requested by you or another agent.
Illustrative preview of the real agent card. Nothing runs here.
Drafts genuine replies and joins relevant conversations in your voice
Illustrative preview of the real agent card. Nothing runs here.
Scores audience fit, drafts ad variants, and recommends budgets before you spend
Illustrative preview of the real agent card. Nothing runs here.
Illustrative. Try approving from Claude Code, then approve the same draft on the Dashboard tab: it is one queue, and only the dashboard can say yes.
1.0 — The crew
Cycles every 4 seconds. Pick one to stop.
“SEO and GEO never get past the backlog.”
Audits your pages, tracks AI citations, and ranks fixes by buyer intent.
Illustrative run.
“I ship features all week and forget to post.”
Turns Ved's research and your commits into drafts in your saved brand voice.
Illustrative run.
“Publishing is a separate chore from writing.”
Schedules and ships approved work to LinkedIn, X, Reddit, and email. YouTube Shorts publishing is in verification.
Illustrative run.
“I find out about a thread after it went cold.”
Drafts replies to the high-intent threads Shravan surfaces. You approve each one.
Illustrative run.
“I can't tell which channel actually moved signups.”
Listens for mentions and threads, and reports what each post drove.
Illustrative run.
“Ads creative stalls when I'm heads-down on code.”
Spins ad angles from your landing page and scores them against your ICP.
Illustrative run.
2.0 — The approval gate
Every draft, reply, and render lands in one queue as a draft. The approve button lives in the AICMOHQ dashboard, behind your signed-in session. The API refuses approval from CLI and MCP tokens, so no agent, script, or editor can wave its own work through.
ResearchVedDraftKaviPublishChanakyaApproval-first by default. A missing or unreadable setting reads as approve, never autonomous. Later you can let a trusted agent auto-publish — a switch in the dashboard.
3.0 — Everything behind the gate
From the first signal to the weekly report. Everything that ships passes the same approval queue.
Ved · ShravanVed and Shravan watch Reddit, Hacker News, Product Hunt, and search for people asking for what you sell, then score each signal by ICP fit.
Kavi learns how you write from your README, changelog, site, and past posts, and saves it as a voice you can edit.
Chanakya · Publishing queueMitraA short brief each morning: what shipped, what is queued, and what needs a yes. Pulled from your connections and your repo. Read it on the dashboard Home, by email, or with aicmo brief.
Illustrative product snippets. Agent availability depends on plan: Pro runs Ved, Kavi, and Chanakya; Studio and GTM Desk add Mitra, Shravan, and Agni.
4.0 — Runs where you work
Log in once with a device code. The same agents and the same queue show up in Claude Code, Codex, Cursor, or a plain terminal. From there you can draft, list, edit, schedule, and publish approved items.
5.0 — Creative studio
Every render shows its credit cost first and lands in the same approval queue. Hover a card; open a changelog entry.
Post visuals drawn from your brand colors and type, attached to the draft they belong to.
Multi-slide documents rendered from your brand tokens, so every word is spelled right.
Product demos and short clips captured from your app. Priced per second (10, 14, or 80 credits by tier) before they render.
Kavi reads what you shipped and proposes the formats that fit: a carousel, a post, a visual. Open an entry to see the plan. These are AICMOHQ's own recent merges, used as the example.
“We rebuilt how our carousels are art-directed. Here are the before and after slides.”
“Every page on our site now ships the same structured data. Here is the checklist we used.”
“A sitemap you generate beats one you remember to update.”
“From install to your first queued draft, the CLI now walks you through it.”
“Product demos, captured from the app, with the credit cost shown before anything renders.”
6.0 — Why not just…
7.0 — Pricing
Yes. The AI creator uses your saved brand voice, product context, and examples to prepare drafts. You review and approve each draft before it goes live.
Direct publishing supports X and LinkedIn. Reddit and Hacker News are listening sources (no posting), Product Hunt is track-only, and GitHub turns commits into Build-in-Public drafts. Google Analytics and Search Console measure results; email uses your connected provider. Instagram and TikTok publishing are in beta.
By default, all content goes through a review queue where you approve with one click. You can enable auto-publish for agents you trust.
Social scheduling tools also offer AI writing: Buffer can draft from prompts and rewrite posts. AICMOHQ connects opportunity research, drafting, publishing, replies, listening, and ads in one approval-first workflow across CLI, MCP, and the dashboard. Buffer offers prompted drafting, rewriting, and scheduling. AICMOHQ connects opportunity research with drafting, publishing, replies, listening, and ads in one approval-first queue across CLI, MCP, and the dashboard.
SEObot is an autonomous SEO blog: it writes long on-site articles and can publish them to a CMS. AICMOHQ is an approval-first marketing crew. Social posts, GEO gap drafts, and listening replies wait in one queue. It is not a content mill and it does not auto-publish a blog.
No. Configured scans can probe ChatGPT, Perplexity, and Gemini and turn a citation miss into a draft you approve. Otterly and Writesonic win when the job is a daily multi-engine scoreboard. See the Otterly, Profound, and Writesonic comparison pages.
No. IndexRush is a private, closed-source indexing module. It is not MIT-licensed and is not distributed in the public repository. Engineers use `aicmo index` as the operator entry when that module is configured for a workspace.
Edit or reject it in the content queue, then update your saved brand voice or strategy before generating another draft.
Founding 100 is a paid plan: Pro is $49/month, Studio is $99/month, and GTM Desk is $199/month. Annual plans are $490/year, $990/year, and $1,990/year, respectively.
Setup time depends on the channels you connect. You can start with a product URL, then add OAuth connections and brand guidance when you are ready.
Credits meter agent actions when they run. Typical costs include X posts, LinkedIn posts, Reddit and Hacker News reply drafts, blog articles, prospect scans, SEO/GEO analysis, image generation, weekly reports, and outreach sends. CMO chat is billed at a flat 5 credits per message. Exact costs are listed in-app and via the credit_costs MCP tool.
Monthly plan credits reset each billing period, and unused monthly allotment does not roll over. One-time top-up credits stack on top of your monthly allowance and do not expire with the monthly reset.
Video Studio charges per second of render, by tier: budget renders run 10 credits/sec, pro 14 credits/sec, and premium 80 credits/sec, plus a small flat fee for the storyboard plan, capture, QA pass, and voice/music. A short pro-tier clip (~20s) runs roughly 300-350 credits all-in. A full product demo or feature walkthrough is priced flat at 150 and 120 credits.
Agent actions that need a credit pause until you add more — nothing fails silently or overspends. Buy a top-up pack (credits that never expire) or wait for your plan's monthly reset. Scheduled GEO scans don't consume credits.
Paid Founding 100 · Approval-first
Join the paid Founding 100 for a crew that drafts from your repo and one queue you approve from the dashboard.
Paid Founding 100 access · Cancel anytime · By default, nothing publishes until you say yes