TRUF at Work 🏗️ What Utility?
TRUF.NETWORK is the decentralized rails for verifiable economic data powering AI and global markets. Data becomes intelligence, intelligence becomes markets, and markets become revenue and the TRUF token is the meter that runs across that stack.
This post explains, in plain terms, what TRUF actually does on the network today and how Truflation Prediction market works with the TRUF.NETWORK
Index publishers pay in TRUF for their index to go live on TRUF. Prediction markets pay in TRUF for market creation.

Truflation launched its Prediction market pilot, which covers MAG‑7 quarterly EPS in a five‑strike, self‑custody event market. Each strike price has a buy and sell market, and each of those markets requires TRUF.

Everything below is how that works underneath.
What TRUF.NETWORK actually is
TRUF.NETWORK is a purpose-built chain for economic data and the markets that reference it. It combines:
- Deterministic SQL execution over Roadrunner consensus - TRUF.NETWORK runs on Roadrunner, a purpose-built two-phase (Propose → Commit) consensus algorithm for database systems. A leader proposes a block, validators vote with an  AppHash  of their post-block state, and the block is committed once a majority (>50%) agree on the same state root. Any node that diverges is halted from consensus. Transactions execute as deterministic SQL against Postgres, so every node arrives at the same answer and the state is cryptographically verifiable and replicable.
- Publishing and reading primitives — data providers publish indices as first-class objects on the network; SDKs in JavaScript, Go, and Python let apps and agents read them.
- Market primitives — create_market is a native transaction type. The SDK and Trufscan explorer read market details straight from those transactions, so every market on the network is inspectable at the protocol layer.
- Order books, settlement, and fees on-chain — the indexer tracks protocol-fee ledgers, discovers the next settlement for every open series, and mirrors order-book fee distributions by the token in which each fee was charged.
- Agent wallets (MAA) — Managed Agent Accounts let a human or an app operate sub-wallets with lifecycle support in the SDKs, so trading bots and data agents can transact under scoped authority.
TRUF is the native gas and access token that ties these pieces together.
Utility 1 — Publishing an index
Every index on TRUF.NETWORK is a live, versioned object with its own address, schema, and update history. Publishing one is not a marketing action; it is an on-chain transaction.
To go live, a publisher pays in TRUF:
- Registration fee — writing the index definition to state.
- Write fees for every update — enforced at the node level, so an index that updates hourly costs more to keep alive than one that updates monthly.
- Insert validation — updates are validated against the index's declared schema, so publishers pay only for updates that pass. The Truflation insert validation alignment work made this uniform across publishers.

This gives TRUF a direct link to real usage: the more indices go live, and the more frequently they update, the more TRUF is consumed. Every one of these transactions is inspectable on the network's own explorer, so publisher activity and protocol-fee flow are auditable in public.

Utility 2 — Creating a market
A market on TRUF.NETWORK, a prediction market, an event market, eventually a perp is a create_market transaction. That transaction requires TRUF.
Every market carries:
- A creation fee in TRUF that mints the market and registers its parameters (underlying, strikes, cutoff, settlement source).
- A settlement binding to a specific data source on the network — the same publishing rail described above — so the outcome resolves against verifiable data rather than a discretionary oracle call.
- An order-book fee schedule whose distributions are mirrored by the indexer and reported in daily totals by token.
Market creators pay TRUF to bring the market into existence. Traders then interact with the order book through agent wallets under standard trading fees.
Trading itself is free. Once a market is live, placing, cancelling, or filling an order costs no gas — a deliberate design choice to keep the order book competitive with centralized venues on execution cost. Settlement happens in USDC, not TRUF: when a market resolves, payouts are made in USDC against the open interest at the time of settlement, with a 2% settlement fee taken from the settled pool.
TRUF is the coordination and creation token; USDC is the trading and payout rail. That split keeps TRUF utility tied to network growth (new indices, new markets) rather than to trading friction.
Utility 3 — The MAG‑7 EPS pilot, concretely
Truflation launched their first live market program as a self-custodial event market on MAG‑7 quarterly EPS. It is the wedge they chose because earnings season is a scheduled, high-attention, well-defined event, and because self-custody removes counterparty risk from the trader's side.
Here is how TRUF shows up across a single ticker:
- One quarterly EPS event per name.
- Five strike prices (outcome bands) per event, each representing a different EPS outcome band.
- Two markets per strike (outcome) — a YES and a NO market — so traders can take either side of every band. Each of those markets is itself a two-sided order book with buys and sells.
- That means ten markets per ticker per quarter, each one a create_market transaction paid in TRUF.

- Across the MAG‑7 — MSFT, AAPL, NVDA, META, AMZN, GOOG, and TSLA — a full earnings cycle spins up seventy markets, all paid in TRUF, before a single trade is placed.

For third parties to quote on these markets TRUF.NETWORK publishes an open-source market-maker bot, an Avellaneda–Stoikov reference implementation that anyone can fork, extend, and run with their own strategies. This code is the simplest working version. Clone it and it will quote on TRUF.NETWORK out of the box. If you want to actually make money doing this, treat it as a foundation and add your own trading strategy on top.
Why this design matters
Two things follow from wiring TRUF into publishing and market creation, rather than bolting it on as a governance token:
- Token demand tracks real product usage. More indices, more updates, more markets, more earnings cycles — each of those is a TRUF transaction. Notably, that demand is decoupled from trading volume — trading is free, settlement is USDC — so TRUF value tracks the supply side of the network: indices, markets, and events. Refill-bot activity, protocol-fee ledgers, and settlement discovery are already observable in treasury tooling and the indexer.
- Every market is anchored to verifiable data. Because publishing and market creation live on the same rails, a market cannot be created against an index that does not exist on the network, and it cannot settle against a source that was not committed to state.
That is the economic model of TRUF.NETWORK, and it is the reason we can move from a self-custodial EPS pilot into CPI perps, dated CPI futures, and further macro instruments on the same primitives.
What comes next
The MAG‑7 EPS pilot is the wedge. The next expansions on the same TRUF rail:
- More earnings names and sectors on the EPS event-market format.
- CPI perps referencing Truflation's real-time CPI, with dated CPI futures and regional CPI perps to follow.
- Third-party index publishers going live on TRUF.NETWORK on the same publishing economics we use ourselves.
If you want to publish an index, create a market, or provide liquidity, the SDKs and Trufscan are open. TRUF is how you plug in.
Sources and further reading: