Skip to content

Trident#

The Trident struct is the main orchestrator for fuzzing operations. It manages the fuzzing client, random number generation, and provides methods for executing transactions and collecting metrics.

pub struct Trident {
    client: TridentSVM,
    fuzzing_data: TridentFuzzingData,
    rng: TridentRng,
}

Core Transaction Processing Methods#

process_transaction#

Processes a transaction with the given instructions and returns the result with logs.

pub fn process_transaction(
    &mut self,
    instructions: &[Instruction],
    log_as: Option<&str>,
) -> TransactionResult

Parameters:

  • instructions - Array of instructions to execute in the transaction
  • log_as - Optional name for the transaction (used in metrics and logging). If None, the transaction will not be logged.

Returns: TransactionResult containing success/failure status and transaction logs.

Description: Executes a transaction containing one or more instructions and returns the results. Use this to test your program's instructions with various inputs. Provide a transaction name to enable metrics collection and logging for that transaction. See TransactionResult for available methods to inspect the transaction outcome.

deploy_program#

Deploys a program to the fuzzing environment.

pub fn deploy_program(&mut self, program: TridentProgram)

Parameters:

  • program - The program binary and metadata to deploy

Description: Makes a program available for testing by deploying it to the fuzzing environment.

deploy_entrypoint (syscall-v2 feature)#

Deploys an entrypoint program to the fuzzing environment.

#[cfg(feature = "syscall-v2")]
pub fn deploy_entrypoint(&mut self, program: TridentEntrypoint)

Parameters:

  • program - The entrypoint program to deploy

Description: Makes an entrypoint program available for testing when using the syscall-v2 feature.

Not Yet Implemented#

get_last_blockhash#

pub fn get_last_blockhash(&self) -> Hash

Status: TODO - Not yet implemented for TridentSVM.

Method Categories#

Specialized Method Groups#

  • Account Management - Account data retrieval, manipulation, and lamport management
  • Time & Clock - Time manipulation and clock-related operations for testing time-dependent logic
  • Random Generation - Generate random data for comprehensive fuzz testing
  • Metrics & Regression - Collect metrics and track account states for regression testing

Native Program Methods#

Feature-Gated Methods#

Getting Started#

For most fuzz tests, you'll primarily use:

  1. process_transaction - Execute your program instructions
  2. Random generation methods - Create varied test inputs
  3. Account management methods - Set up and inspect account states

Example Usage#

use trident_fuzz::*;

#[flow]
fn basic_fuzz_example(&mut self) {
    // Generate random test data
    let amount = self.random_from_range(1..1000u64);
    let recipient = self.random_pubkey();

    // Create and execute transaction with logging
    let transfer_ix = self.transfer(&self.payer().pubkey(), &recipient, amount);
    let result = self.process_transaction(&[transfer_ix], Some("Transfer Test"));

    // Verify results
    assert!(result.is_success());

    // Or execute without logging (pass None)
    let result = self.process_transaction(&[transfer_ix], None);
}