api_reference.md

API Reference

Overview

This document provides comprehensive API reference for VectorSmuggle modules and components.

Core Modules

Configuration (config.py)

Classes

Config

Main configuration class with environment variable support.

from config import Config

config = Config()

SteganographyConfig

Configuration for steganographic techniques.

EvasionConfig

Configuration for evasion methods.

QueryConfig

Configuration for query enhancement features.

Steganography Module

steganography.obfuscation

EmbeddingObfuscator

Applies various obfuscation techniques to embeddings.

from steganography.obfuscation import EmbeddingObfuscator

obfuscator = EmbeddingObfuscator(noise_level=0.01)
obfuscated = obfuscator.obfuscate(embeddings, techniques=["noise", "rotation"])

MultiModelFragmenter

Fragments data across multiple embedding models.

steganography.timing

TimingController

Controls timing patterns for covert operations.

Document Loaders

loaders.document_factory

DocumentLoaderFactory

Factory for creating appropriate document loaders.

from loaders.document_factory import DocumentLoaderFactory

factory = DocumentLoaderFactory()
documents = factory.load_documents(file_paths)

Evasion Module

evasion.behavioral_camouflage

BehavioralCamouflage

Implements behavioral camouflage techniques.

evasion.traffic_mimicry

TrafficMimicry

Mimics legitimate traffic patterns.

Query Enhancement

query.advanced_engine

AdvancedQueryEngine

Enhanced query engine with multiple strategies.

query.context_reconstruction

ContextReconstructor

Reconstructs document context from fragments.

Analysis Tools

analysis.risk_assessment

VectorExfiltrationRiskAssessor

Comprehensive risk assessment for vector exfiltration.

analysis.forensic_tools

EvidenceCollector

Collects digital evidence from vector stores.

Error Handling

All modules implement comprehensive error handling with custom exceptions:

• ConfigurationError: Configuration-related errors
• SteganographyError: Steganography operation errors
• EvasionError: Evasion technique errors
• QueryError: Query processing errors

Logging

All modules use structured logging with configurable levels:

import logging

logger = logging.getLogger(__name__)
logger.info("Operation completed", extra={"operation": "embed", "status": "success"})

Type Hints

All public APIs include comprehensive type hints for better IDE support and code clarity.

Examples

See individual module documentation and the main README for usage examples.