Introduction

Pina is a Solana program framework built on pinocchio. Programs written with it deploy with the smallest binaries of any Solana framework, spend compute like hand-written code, and evolve through managed ABI migrations. Every on-chain crate is no_std and allocation-free.

This book is the documentation home for the project. It explains the ideas behind the framework, walks through building and shipping a program, and records the trade-offs the codebase is built on. API reference lives on docs.rs.

Pina is unaudited and still hardening. Read the security model and the production-readiness gate before you deploy a program that holds value.

Where to start

• New to Pina? Follow Getting Started, then build your first program end to end.
• Coming from Anchor? Migrating from Anchor maps each Anchor concept to its Pina equivalent.
• Evaluating frameworks? The framework comparison measures size and compute units for the same programs written in five frameworks.
• Shipping an upgrade? How ABI migrations flow covers version envelopes, transitions, and the publication ledger.

What this book covers

• Start here: the project goals, toolchain setup, tutorials, and the annotated examples.
• Framework: core concepts (#[account], #[instruction], #[derive(Accounts)], discriminators, and validation chains), declarative validation, crate features, and the Codama workflow for generated clients.
• Tooling: the complete CLI reference, including automation-safe output contracts, and the packaged agent skill.
• ABI migrations: how account and instruction schemas evolve without breaking deployed programs, and a migration guide for every breaking release.
• Performance: compute-unit and program-size guidance, backed by the benchmarks in CI.
• Security: the security model, the PinaPod safety boundary, and the security lints that check program source.
• Project: architecture decision records, the development workflow, and CI and releases.