Threat Model

Understanding the security assumptions and threats

Draft Documentation

This documentation is under development and may be incomplete or subject to change.

Overview

This document describes the threat model for ComputeNet, including security assumptions, potential attack vectors, and mitigation strategies.

Security Goals

ComputeNet aims to provide the following security properties:

  • Correctness — Verified results are computationally correct
  • Integrity — Results cannot be tampered with
  • Availability — The network remains operational
  • Censorship resistance — Jobs cannot be arbitrarily blocked

Trust Assumptions

The protocol makes the following assumptions:

  • At least 2/3 of validator stake is honest
  • Cryptographic primitives are secure
  • Network can eventually deliver messages
  • Clients can verify proofs correctly

Threat Categories

Validator Attacks

  • Incorrect computation with false proofs
  • Collusion between validators
  • Selective job execution
  • Result manipulation before attestation

Network Attacks

  • Eclipse attacks on individual nodes
  • Network partitioning
  • DDoS against validators or coordinators
  • Message replay or reordering

Cryptographic Attacks

  • Proof forgery attempts
  • Signature manipulation
  • Hash collision exploitation
  • Side-channel attacks on proof generation

Economic Attacks

  • Stake manipulation
  • Bribery of validators
  • Front-running of job results
  • Resource exhaustion attacks

Mitigations

Key mitigation strategies include:

  • Cryptographic proofs for all computations
  • Economic penalties (slashing) for misbehavior
  • Multi-validator attestation requirements
  • Randomized validator selection
  • Rate limiting and resource controls

Draft Analysis

The threat model is under active review. Additional attack vectors and mitigations may be identified through security analysis.