Security

Security documentation for the Seesaw protocol.

Overview

Seesaw is designed with security as a core principle:

• Permissionless: No trusted parties required
• Trustless: Protocol enforces all rules on-chain
• Audited: Security review completed
• Open Source: Publicly verifiable

In This Section

• Threat Model - Attack surfaces and mitigations
• Invariants - Protocol guarantees

Key Security Properties

Trust Assumptions

Security Measures

Arithmetic Safety

// All arithmetic uses checked operations
let result = a.checked_add(b).ok_or(Error::Overflow)?;
let product = x.checked_mul(y).ok_or(Error::Overflow)?;

Account Validation

// Validate ownership, type, and PDA derivation
validate_owner(account, &program_id)?;
validate_discriminator::<MarketAccount>(account)?;
validate_pda(account, seeds, bump, &program_id)?;

State Machine

// Enforce valid state transitions
require!(
    can_transition(current_state, instruction),
    Error::InvalidStateTransition
);

Reporting Vulnerabilities

Responsible Disclosure

If you discover a vulnerability:

1. DO NOT exploit or publicize
2. Report via security@seesaw.markets
3. Include detailed reproduction steps
4. Allow 90 days for fix

Bug Bounty

Scope

In Scope:

• Core protocol smart contract
• Oracle integration
• Order book logic
• Settlement logic

Out of Scope:

• Frontend applications
• Third-party integrations
• Already known issues

Audit Status

January 2026 Audit

See Threat Model for details.

Next Steps

• Review Threat Model
• Understand Invariants