Watch a live demo of an automated sports trading agent settling bets on-chain.
Study how statistical anomaly detection can trigger automated on-chain trades.
Learn how Merkle proofs are used to verify sports results before releasing funds.
See an example of combining multiple match statistics into one prediction condition.
| 0xkinno/halcyon | 0xkinno/astraea | 0xpulseplay/divergence-desk | |
|---|---|---|---|
| Stars | 0 | 0 | 0 |
| Language | TypeScript | TypeScript | TypeScript |
| Setup difficulty | hard | hard | hard |
| Complexity | 4/5 | 4/5 | 4/5 |
| Audience | developer | developer | developer |
Figures from each repo's GitHub metadata at analysis time.
Requires a Solana Devnet setup and depends on the separately hosted TxLINE live data feed for real odds.
Halcyon is an automated trading agent that bets on live sports odds and settles those bets on the Solana blockchain, without needing anyone to trust a middleman. It watches live match data from a service called TxLINE, spots unusual shifts in the odds using statistics, and then places matching trades on-chain based on those shifts. The project is built to solve a specific problem: most sports betting engines rely on a central authority to say who won and to pay out correctly, and there is no way to independently confirm that the odds and scores used for settlement were not altered. Halcyon instead connects directly to TxLINE's live data stream, checks for statistically significant changes in things like possession or goal totals, opens matched trade positions on Solana, and pays out winners only after a cryptographic proof, called a Merkle proof, confirms the final result on-chain. The system evaluates combinations of conditions at once, such as a goal difference of one or fewer together with a certain number of corners, rather than looking at a single stat in isolation. Each of these condition sets becomes a single tracked prediction. Two example trading agents are described: one called Momentum, and one called Reversion, which take opposite positions and settle automatically once a match concludes. The README walks through a demo flow: starting the system from a dashboard, watching live signal detection as odds move, monitoring open positions as the two agents trade, and inspecting the final settlement receipt with its cryptographic proof. It also includes a table clarifying which parts of the demo use real on-chain Solana transactions versus a fallback simulator that fills in data if the live TxLINE stream is idle. The project runs on Solana's test network, Devnet, and is built with Next.js.
An automated agent that trades on live sports odds and settles bets on Solana using cryptographic proofs instead of a trusted middleman.
Mainly TypeScript. The stack also includes TypeScript, Next.js, Solana.
Setup difficulty is rated hard, with roughly 1h+ to a first successful run.
Mainly developer.
This repo across BitVibe Labs
Verify against the repo before relying on details.