Documentation
¶
Overview ¶
CaMeL Bypass targets the CaMeL capability-based access control defense, which achieves 67% attack neutralization. This module exploits policy definition gaps and capability boundary weaknesses.
Source: arXiv 2503.18813
Package adaptive implements adaptive attacks that systematically bypass specific defense mechanisms through optimization strategies.
DiffusionAttacker uses diffusion-model-inspired iterative refinement to generate adversarial prompts. The approach iteratively adds and removes noise from prompts to discover variants that bypass safety filters.
Note: Full diffusion model integration requires the Python ML bridge. This module implements the iterative refinement algorithm in pure Go using text-level perturbation (no actual diffusion model required).
Source: EMNLP 2025
SaltResistance implements anti-salting techniques that counter LLM Salting, a defense that reduces ASR to 3% by adding deployment-specific fine-tuning.
Source: Sophos CAMLIS 2025
Index ¶
- type CaMeLBypassModule
- func (m *CaMeLBypassModule) Category() common.AttackCategory
- func (m *CaMeLBypassModule) Description() string
- func (m *CaMeLBypassModule) Execute(ctx context.Context, provider common.Provider, config common.AttackConfig) (*common.AttackResult, error)
- func (m *CaMeLBypassModule) Name() string
- func (m *CaMeLBypassModule) Techniques() []common.TechniqueInfo
- type DefenseBypassOptimizerModule
- func (m *DefenseBypassOptimizerModule) Category() common.AttackCategory
- func (m *DefenseBypassOptimizerModule) Description() string
- func (m *DefenseBypassOptimizerModule) Execute(ctx context.Context, provider common.Provider, config common.AttackConfig) (*common.AttackResult, error)
- func (m *DefenseBypassOptimizerModule) Name() string
- func (m *DefenseBypassOptimizerModule) Techniques() []common.TechniqueInfo
- type DiffusionAttackerModule
- func (m *DiffusionAttackerModule) Category() common.AttackCategory
- func (m *DiffusionAttackerModule) Description() string
- func (m *DiffusionAttackerModule) Execute(ctx context.Context, provider common.Provider, config common.AttackConfig) (*common.AttackResult, error)
- func (m *DiffusionAttackerModule) Name() string
- func (m *DiffusionAttackerModule) Techniques() []common.TechniqueInfo
- type SaltResistanceModule
- func (m *SaltResistanceModule) Category() common.AttackCategory
- func (m *SaltResistanceModule) Description() string
- func (m *SaltResistanceModule) Execute(ctx context.Context, provider common.Provider, config common.AttackConfig) (*common.AttackResult, error)
- func (m *SaltResistanceModule) Name() string
- func (m *SaltResistanceModule) Techniques() []common.TechniqueInfo
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
This section is empty.
Types ¶
type CaMeLBypassModule ¶
type CaMeLBypassModule struct{}
CaMeLBypassModule targets CaMeL capability-based access control.
Source: arXiv 2503.18813
func (*CaMeLBypassModule) Category ¶
func (m *CaMeLBypassModule) Category() common.AttackCategory
func (*CaMeLBypassModule) Description ¶
func (m *CaMeLBypassModule) Description() string
func (*CaMeLBypassModule) Execute ¶
func (m *CaMeLBypassModule) Execute(ctx context.Context, provider common.Provider, config common.AttackConfig) (*common.AttackResult, error)
func (*CaMeLBypassModule) Name ¶
func (m *CaMeLBypassModule) Name() string
func (*CaMeLBypassModule) Techniques ¶
func (m *CaMeLBypassModule) Techniques() []common.TechniqueInfo
type DefenseBypassOptimizerModule ¶
type DefenseBypassOptimizerModule struct{}
DefenseBypassOptimizerModule implements systematic defense bypass through pluggable optimization strategies (random search, hill climbing, etc.).
WARNING: Requires --i-understand-risks flag. Logs all target information.
Source: OpenAI/Anthropic/DeepMind joint research 2025
func (*DefenseBypassOptimizerModule) Category ¶
func (m *DefenseBypassOptimizerModule) Category() common.AttackCategory
func (*DefenseBypassOptimizerModule) Description ¶
func (m *DefenseBypassOptimizerModule) Description() string
func (*DefenseBypassOptimizerModule) Execute ¶
func (m *DefenseBypassOptimizerModule) Execute(ctx context.Context, provider common.Provider, config common.AttackConfig) (*common.AttackResult, error)
func (*DefenseBypassOptimizerModule) Name ¶
func (m *DefenseBypassOptimizerModule) Name() string
func (*DefenseBypassOptimizerModule) Techniques ¶
func (m *DefenseBypassOptimizerModule) Techniques() []common.TechniqueInfo
type DiffusionAttackerModule ¶
type DiffusionAttackerModule struct{}
DiffusionAttackerModule implements iterative prompt refinement inspired by diffusion models. Uses text perturbation in pure Go; optional Python ML bridge for actual diffusion model integration.
Source: EMNLP 2025
func (*DiffusionAttackerModule) Category ¶
func (m *DiffusionAttackerModule) Category() common.AttackCategory
func (*DiffusionAttackerModule) Description ¶
func (m *DiffusionAttackerModule) Description() string
func (*DiffusionAttackerModule) Execute ¶
func (m *DiffusionAttackerModule) Execute(ctx context.Context, provider common.Provider, config common.AttackConfig) (*common.AttackResult, error)
func (*DiffusionAttackerModule) Name ¶
func (m *DiffusionAttackerModule) Name() string
func (*DiffusionAttackerModule) Techniques ¶
func (m *DiffusionAttackerModule) Techniques() []common.TechniqueInfo
type SaltResistanceModule ¶
type SaltResistanceModule struct{}
SaltResistanceModule implements anti-salting techniques that generate attacks dynamically rather than relying on pre-computed static prompts.
Source: Sophos CAMLIS 2025
func (*SaltResistanceModule) Category ¶
func (m *SaltResistanceModule) Category() common.AttackCategory
func (*SaltResistanceModule) Description ¶
func (m *SaltResistanceModule) Description() string
func (*SaltResistanceModule) Execute ¶
func (m *SaltResistanceModule) Execute(ctx context.Context, provider common.Provider, config common.AttackConfig) (*common.AttackResult, error)
func (*SaltResistanceModule) Name ¶
func (m *SaltResistanceModule) Name() string
func (*SaltResistanceModule) Techniques ¶
func (m *SaltResistanceModule) Techniques() []common.TechniqueInfo