Hyperstack
Hyperstack provides real-time streaming data from Solana programs via WebSocket. Think of it as a typed, structured pub/sub system for on-chain state.
Key concepts:
- Stack — a deployed streaming server that watches Solana programs and produces structured entity updates
- Entity — a domain object (e.g.
OreRound,OreMiner) with typed fields organized in sections - View — a lens on an entity:
state(keyed lookup),list(collection), or custom (sorted/filtered) - SDK — TypeScript, React, or Rust client libraries that connect to a stack's WebSocket and provide typed access to views
Which Skill Do You Need?
Determine the user's intent, then load the appropriate sub-skill. Both sub-skills handle their own prerequisites, CLI setup, and schema discovery.
Consuming an existing stack (connecting to streams, subscribing to data, building UIs): → Load the hyperstack-consume skill
Building a custom stack (writing Rust DSL, defining entities from program IDLs, deploying): → Load the hyperstack-build skill
Not sure yet? Ask the user whether they want to consume data from an existing stack or build a new one from a Solana program IDL.
Quick Reference
Scaffold a new app: npx hyperstack-cli create my-app
CLI install: cargo install hyperstack-cli (binary: hs) or npm install -g hyperstack-cli (binary: hyperstack-cli). Prefer hs.
Discover available stacks: hs explore --json
Current public stacks: ore (ORE mining rounds, miners, treasury state). Prepackaged SDK types: npm install hyperstack-stacks