Building a Program with Pinocchio

A Pinocchio program combines a thin entrypoint, strict account validation, Borsh or opcode dispatch, and CPI helpers - mirroring native structure with less boilerplate overhead.

Recipe

entrypoint!(process_instruction);
fn process_instruction(id: &Pubkey, accounts: &[AccountInfo], data: &[u8]) -> ProgramResult {
    match data.first() { Some(0) => init(id, accounts), _ => Err(ProgramError::InvalidInstructionData) }
}

When to reach for this:

• Greenfield performance-sensitive program.
• Rewriting one Anchor instruction.
• Learning lightweight framework patterns.

Working Example

pub fn init(program_id: &Pubkey, accounts: &[AccountInfo]) -> ProgramResult {
    let [payer, state, system @ ..] = accounts else { return Err(ProgramError::NotEnoughAccountKeys) };
    if !payer.is_signer() { return Err(ProgramError::MissingRequiredSignature); }
    // CPI create via pinocchio invoke / system instructions
    Ok(())
}

What this demonstrates:

• Fixed account slice destructuring.
• Signer verified first.
• System CPI for create follows validation.

Deep Dive

Module Layout

• entry.rs / lib.rs
• instructions/
• state.rs
• cpi.rs

IDL

Generate with Codama for @solana/kit 7.0.0 clients.

Rust Notes

// Centralize account validation helpers.

Gotchas

• Skip validation - Worse than Anchor - no guardrails. Fix: Copy native checklists.
• No IDL published - Clients break. Fix: Codama pipeline.
• Mixed pinocchio/solana-program types - FFI confusion. Fix: Pick one SDK surface per crate.
• Feature creep - Re-add Anchor-like macros. Fix: Stay disciplined.
• No tests - CU regressions unnoticed. Fix: LiteSVM CI.

Alternatives

FAQs

Related

• Pinocchio Basics - Intro
• Native Program Basics - Parallel concepts
• Interop & Migration - Moving stacks

Stack versions: This page was written for Agave 4.1.1, Solana CLI 3.0.10, Anchor 0.32.1, anchor-lang 0.32.1, Rust 1.91.1, @solana/kit 7.0.0, Surfpool 0.12.0, and LiteSVM 0.6.x.