Skip to content

API Reference

Complete API documentation for the Empathy Framework.

Overview

The Empathy Framework provides a comprehensive Python API for building AI systems with five levels of empathy:

  • Level 1: Reactive (basic Q&A)
  • Level 2: Guided (clarifying questions)
  • Level 3: Proactive (suggests improvements)
  • Level 4: Anticipatory (predicts problems)
  • Level 5: Transformative (reshapes workflows)

Core Modules

EmpathyOS

Main entry point for the framework. Handles interaction logic, level progression, and trust management.

Key Classes: - EmpathyOS - Primary interface for empathy interactions

Configuration

Configuration management for the framework.

Key Classes: - EmpathyConfig - Configuration container with validation - load_config() - Load configuration from files or environment

Core

Core data structures and state management.

Key Classes: - CollaborationState - Tracks trust, level, and interaction history - EmpathyResponse - Response container with metadata - EmpathyLevel - Enumeration of empathy levels

Pattern Library

Pattern recognition and learning system for multi-agent coordination.

Key Classes: - PatternLibrary - Manages pattern discovery and sharing - Pattern - Individual pattern with confidence tracking - PatternMatch - Pattern matching results

Persistence

Data persistence for patterns, metrics, and state.

Key Classes: - PatternPersistence - Save/load pattern libraries (JSON, SQLite) - StateManager - Manage user collaboration states - MetricsCollector - Track usage metrics and performance

LLM Toolkit

LLM provider integration with security controls.

Key Classes: - EmpathyLLM - Unified LLM interface with empathy integration - PIIScrubber - PII detection and scrubbing - SecretsDetector - API key and credential detection - AuditLogger - Compliance and security audit logging

Installation

pip install empathy-framework

For LLM support: 

pip install empathy-framework[llm]

For healthcare applications: 

pip install empathy-framework[healthcare]

Basic Usage

from empathy_os import EmpathyOS

# Create Level 4 (Anticipatory) system
empathy = EmpathyOS(
    user_id="user_123",
    target_level=4,
    confidence_threshold=0.75
)

# Interact
response = empathy.interact(
    user_id="user_123",
    user_input="I'm about to deploy this change to production",
    context={"deployment": "production"}
)

print(response.response)
print(f"Level: {response.level}")
print(f"Predictions: {response.predictions}")

License

Fair Source License 0.9 - Free for teams up to 5, commercial license required for 6+ employees.