beyond the demo

the docs, and what happens next

herit was built in a hackathon, but the problem it picks is not a hackathon problem. this page is the honest version of both: exactly what exists today and where to read it, then what it would take to make this something a family could actually rely on — and how it would find the people who need it.

read the repo see the deployed contracts

documentation

read the source, not the pitch

everything herit claims on the landing page is checkable. these are the documents it is checkable against.

start here

ARCHITECTURE.md

the design document. why proof of life is a Selfie Check and not a timestamp, what each of the five contracts owns, the four key flows, and the security assumptions behind the backend attestor.

readme

README.md

the working notes this project started from — the original problem statement and the open questions, including the ones we deliberately left unanswered for the hackathon.

addresses

documents/deployments.md

every address and transaction hash on Sepolia: the five herit contracts, the ENS layer under herit.eth, and the frozen ENSv2 hackathon set they are wired to.

indexing

frontend/subgraph

the subgraph that feeds the live dashboard — schema, manifest and mappings for every event the five contracts emit.

contracts

src/

the Solidity itself. a Foundry project whose five contracts deploy as one nonce sequence, all verified on Sepolia Etherscan.

github

the repository

everything above in one place, including the Foundry tests, the deploy scripts and the checkpoint guides recording how each piece was brought up.

what it is built on

ENSv2: heir subnames, and the roles that gate them
World ID: Selfie Check liveness and uniqueness
The Graph: indexes every herit event — the live dashboard and the estate explorer read from it
Foundry: contracts, tests and deploy scripts
Next.js: this app, with wagmi and viem underneath

roadmap

how a demo becomes something you would put real money behind

phase 0 is on Sepolia now and can be verified line by line. everything after it is stated as intent, in the order the constraints actually lift.

phase 0

the hackathon build

shipped

five contracts live and verified on Sepolia, deployed as one set
an ENSv2 registry per estate, heirs minted as subnames carrying relationship and share records
ROLE_HEIR_CLAIM withheld at registration and granted on unlock — the withheld bit is the inheritance
Selfie Check on both sides: the grantor's recurring check-in and the heir's one-time claim
a subgraph indexing every event, so the dashboard and the public estate explorer are live rather than mocked

phase 1

from demo to something worth trusting

next three months

a reminder service — email, push and World App notification before a window closes. the contract stays the source of truth; a missed reminder must never be the reason an estate unlocks
Safe{Wallet} module integration, so an estate can be a grantor's existing smart account rather than a separate willed deposit. this is the real path to inheriting a whole wallet
a guardian committee: a few grantor-nominated, World ID-verified humans who can veto an unlock inside a time-boxed window, never indefinitely
an external audit of the five contracts, and a public bug bounty, before any mainnet funds
the World App mini app shell, so the whole flow runs where verified humans already are

phase 2

mainnet, and removing ourselves from the trust path

when ENSv2 reaches mainnet

on-chain World ID verification once one chain hosts both the verifier and ENSv2, retiring the backend attestor as a trust assumption
multi-asset and multi-chain vaults — NFTs, LP positions, and estates that span more than one chain
per-heir rules beyond a flat share: token-scoped allocations, staged release, and heirs who are themselves estates
a liveness cadence tuned to risk, so a large estate checks in more often than a small one

phase 3

the part that is a business, not a protocol

later

a legal wrapper — pairing the on-chain estate with a jurisdiction-aware document, so an unlock is recognised off-chain too
an estate-planner and fiduciary channel: the professionals families already trust with this decision
on-chain insurance — cover a grantor can buy against a wrongful unlock, underwritten against the check-in record the contracts already keep, so a family is made whole even when the timer was the thing that failed
custodian and exchange integrations, so assets that are not self-custodied can still name an heir

and how it pays for itself

opening an estate stays free — a product nobody has heard of cannot also charge at the door. revenue comes from what only matters once the product has worked: a basis-point fee on assets released at claim, and a subscription for the reminder and guardian services that make missing a window unlikely in the first place.

go to market

nobody searches for this. so we go to them

inheritance is bought in a moment, not browsed for. every channel below is picked because it reaches someone already standing in that moment.

distribute inside World App

herit's core requirement — a verified, living human — is the one thing every World App user has already satisfied. shipping as a mini app puts the product in front of an audience pre-qualified for the exact primitive it depends on, with no wallet to install and no gas to explain.

World App mini app store

start with people who already own a name

an ENS holder has already decided their on-chain identity is worth paying for. for them herit is one more subname under a name they own, not a new account — the shortest distance between hearing the pitch and signing the first transaction.

ENS community and name-holder outreach

be a feature inside wallets, not only a destination

nobody wakes up wanting to do estate planning. it becomes real in the moment someone is already looking at their balance. a “name an heir” entry point inside wallets and Safe apps meets that moment, and makes herit distribution-led rather than search-led.

wallet and Safe app integrations

sell the problem, not the mechanism

the market is not people searching for ENS subname inheritance. it is people who have read one more story about coins buried with their owner. writing plainly about that — and about why a dead-man's switch a bot can ping is worse than nothing — is what turns a stranger into someone who opens an estate.

content, post-mortems, founder communities

go where the loss is already priced

DAO treasuries with one signer, funds holding keys on behalf of other people, and long-term holders with no succession plan all already treat this as an unpriced risk. they are a small, reachable, high-intent set of first customers.

direct outreach to DAOs and funds