One source in, a full suite out, in your voice and not in AI's.
The In-Voice Content Engine takes one thing you already have (a paper, a post, a rough note, a link) and turns it into a real blog post plus platform-specific social variants, every piece conditioned to your brand voice and checked against it before it ships. Not a generic draft you then have to rewrite so it stops sounding like a robot, a suite that already sounds like you.
Turn one source into a full suite See a sample suite
THE PROBLEM
Publishing consistently is a grind, and the tools meant to fix it made it worse. Everyone now has a button that spits out a confident, fluent, completely generic draft, so the feeds are full of the same flat AI voice and none of it sounds like a person. You are left doing the part the tool skipped, rewriting every line until it sounds like your brand again, which is most of the work. And if you are in a regulated or voice-sensitive space, one off-voice or unsubstantiated sentence going out under your name is not a small problem. So you get the worst trade, a draft fast enough to tempt you and wrong enough that you cannot ship it as-is.
HOW IT WORKS
The engine runs the same passes a good writer does, in order. First it reads your source and pulls the actual angle out of it (the narrative, the emotional temperature, the strongest hook), so the piece is built around something real and not a summary of nothing. Then it writes the full blog post. Then, and this is the part that matters, it conditions that draft to your brand voice with Voice V3, and checks it, so you get back not just the conditioned text but a count of what it fixed and which voice rules it applied. Only after the piece sounds like you does it fan out into platform-specific variants for LinkedIn, X, Bluesky, and Threads, each one shaped for that platform instead of the same paragraph pasted four times.
WHAT YOU GET
From one source, in one run: - a full blog post, built around the real hook it found in your source, not a bland recap, - your voice, applied and checked, so the draft comes back conditioned to your brand with a plain report of what it changed and which voice systems it ran, - social variants per platform (LinkedIn, X, Bluesky, Threads), each shaped for where it is going, - a suite you can actually ship, or hand to a human for a final look, instead of a first draft you have to rebuild.
WHO IT'S FOR
Anyone who has to publish across channels and cannot afford to sound like everyone else. Solo founders and small marketing teams who are the whole content department. Agencies keeping several distinct brand voices straight without blurring them into one. Voice-sensitive and regulated brands that need every line to sound right and stand up. If you have ever generated a draft and then spent longer fixing the voice than it would have taken to write it, this is for you.
PRICING (the ladder)
- Free / dogfood: the internal content engine we run on our own blog and social. This is literally how our posts get drafted. - Solo: one brand voice, one source in, a full blog-plus-social suite out, conditioned and checked. - Team: multiple brand voices (one per account), calendar scheduling, and distribution hooks. - Verified tier: the claims upgrade, where every factual line in the content is checked against an approved-claims set with a correction pass, so a regulated brand cannot ship an unsubstantiated statement. This rides the same verification discipline that is the core of what we build.
IP is licensed, never assigned. The engine stays ours, the content is yours.
THE PROOF (dogfood)
We draft our own content with this, every week. The founder's blog posts and multi-platform responses come out of the same content suite, conditioned to the founder's real voice (the commas and parentheticals, no em-dashes, none of the usual AI tells), and checked before anything is queued. This very page was drafted in-voice by the engine it describes. The product's first user is the company that builds it, so we are not selling you a voice engine we do not run on ourselves.
HONEST NOTE
Straight facts about where this is. The engine is built, and it is built the honest way, it reuses capabilities we already had (the source analysis, the text generation, the Voice V3 conditioning, the social-variant generation) rather than a fresh pile of code, and the deliverable passes our quality gate. What it does not yet have is a long public record of live runs for an outside brand, because that needs a running system, a funded model key, and your own voice profile loaded, and those are switches we have not thrown for an external user yet. Two more honest lines. The voice conditioning cuts the AI tells down hard, it does not promise zero, so a human look before you publish stays in the loop (that is by design, and it is its own part of the roadmap). And the claims-checking that keeps you from shipping an unsubstantiated line is the verified tier on the ladder, it is not the free suite. We would rather tell you that than pretend the first draft is already the audited one.
Turn one source into a full suite
*This page is a specification. The capability it describes is not built yet, and nothing here is a claim that it runs today.*