Security Model¶
⚠️ STUB DOCUMENTATION: This file contains placeholder content and needs to be expanded with detailed security model documentation.
Security mechanisms and protections in the Verdikta Dispatcher system.
Commit-Reveal Scheme¶
Purpose¶
Prevents oracles from seeing other responses before submitting their own, reducing bias and collusion.
Process¶
- Commit Phase: Oracle submits hash of their response
- Reveal Phase: Oracle reveals the actual response
- Verification: System verifies response matches commit
Oracle Authorization¶
Operator Restrictions¶
Only authorized contracts can request evaluations through the ArbiterOperator.
Reputation Requirements¶
Oracles must maintain minimum reputation scores to participate.
Economic Security¶
Stake Requirements¶
Oracles must stake tokens that can be slashed for malicious behavior.
Fee Escrow¶
Fees are held in escrow until successful completion.
Request Validation¶
Input Sanitization¶
All inputs are validated to prevent injection attacks.
Rate Limiting¶
Protection against spam and denial-of-service attacks.
Audit Status¶
The smart contracts have been audited for security vulnerabilities. Audit reports are available in our GitHub repository.
Best Practices¶
For secure integration patterns: - Integration Examples - Error Handling - Troubleshooting Guide