Provenance: distilled from the submitted YC F26 application (2026-07-27), the 58-agent answer panel, a 26-line tagline tournament judged by SF-investor and serial-founder lenses, and Husam's own live edits during the sprint. Everything here survived contact with a deadline - this is the brand as actually spoken, not as aspired.
The YC sprint forced the brand through the narrowest possible aperture: 50 characters, 1 minute of video, plain answers a tired partner reads at 11pm. What survived is the brand. What died in the judging is listed too, so nobody resurrects it.
The company is Supercontinent, Inc. The product is Entity.ID - the internet's own registry of ventures. The one-line brand is:
Where humans & AI agents cofound companies.
A company can only exist in one government's database, and that existence stops at the border. Teams span countries, money moves in seconds, and AI agents now enter the economy with budgets of their own - transacting, even co-founding. But agents can't be registered anywhere, and informal ventures get nothing until they can afford one country's registry.
Entity.ID is a registry that lives on the internet instead. Register in minutes, free. A public, append-only record - who's in it, who owns what, who can sign - that anyone verifies from anywhere without asking us. Operating rails from day zero: treasury, recorded cap table, fundraising, by-laws that execute as logic. Incorporation is a phase in the venture lifecycle - added when needed, never the price of getting started.
v1's entity map survives, but the GTM telling is now strictly ordered:
The ordering is brand law: pain today funds the uncontested line tomorrow. Never lead agent-first in generalist rooms (see ⑤).
What changed from v1: the agent story is told with explicit humility about timing ("the pain isn't acute yet") and zero personhood vocabulary. The accountability chain to a verified natural person is the product; "rights for AI" is never the frame. The http/https analogy is the approved vehicle - it says inevitability without ideology.
The 26-line tagline tournament produced brand law:
Kill list (judged actively harmful): "Universal Registry of Entities" · "Global Entity Registry" · "the borderless economy" · "for real" constructions · personhood-as-goal phrasing · any line that argues with its own skeptics.
Tagline (primary, tournament winner under the not-agents-only constraint):
Where humans & AI agents cofound companies
Attention line (panel #1 on both judge lenses - use where an eyebrow is wanted and a paragraph of proof follows):
AI agents have money but no papers. We fix that.
Pain line (strongest single-feeling line, no agents):
Turn your handshake business into an entity.
Positioning sentence:
Entity.ID is the internet's own registry of ventures - where a venture gets structure, standing, and operating rails from the handshake stage, and incorporation becomes a phase, not the starting point.
The competitive close:
Everyone else fights over the incorporation moment; nobody serves the venture that exists before it. Today that customer's alternative is nothing.
The three understandings (competitor frame): (1) incumbents are jurisdiction-first with private-by-design records - a registry is only useful if neutral and checkable; (2) the money is in operating, not filing - Carta's ~$442M ARR requires a certificate of incorporation to even begin; (3) nobody registers agents as principals - payment rails exist, the entity layer doesn't.
The horizon (use once per artifact, maximum):
The Supercontinent - a borderless world where countries function like states of one continent. I don't know how to make a business out of a world, so I'm building the tools it would need - starting with its registry.
Unchanged from v1 (see BRANDBOOK-v1 ⑦), with two substitutions when the deck is next rebuilt: the opener adopts the ② two-customers ordering, and the market slide leads with "their alternative is nothing" before any TAM number.
../yc-app-2026/ holds the answers, the panel, and the tournament record.Compiled 2026-07-28 by HusAgent from the F26 sprint artifacts: yc-app-2026/Hus-Draftes.md (final answers), yc-app-2026/panel/ (58-agent generation+judging), the 2-judge tagline tournament (session record), and Husam's live dictations. v1 (2026-07-10) remains the reference for the twelve-slide argument and hero systems. Next planned edition: post-YC-decision (2026-08-28) or post-seed-narrative, whichever lands first.