Examples¶

Practical examples demonstrating ProactiveAgent features. All source code is available in the examples directory.

Basic Usage¶

Example	Description
minimal_chat.py	Minimal implementation demonstrating core agent initialization, message handling, and default AI-based decision making.
beautiful_chat.py	Terminal interface with rich formatting, colored output, and visual feedback for agent decisions.
minimal_callbacks.py	Complete callback implementation showing response, sleep time, and decision callbacks for monitoring agent behavior.

Configuration¶

Example	Description
all_config_parameters.py	Comprehensive reference documenting all available configuration parameters with inline explanations and default values.

Decision Engines¶

Decision engines evaluate conversation state and determine whether the agent should respond.

Example	Implementation
ai_based_decision_engine.py	LLM-powered decision making that evaluates context, engagement, and conversation flow.
threshold_decision_engine.py	Rule-based system using configurable thresholds for priority levels and engagement metrics.
function_based_decision_engine.py	Custom decision logic implemented as a Python function with access to message history and context.
simple_decision_engine.py	Always-respond implementation useful for testing and debugging agent responses.
custom_decision_engine.py	Full custom implementation extending the base `DecisionEngine` class for complex logic requiring state management.

Sleep Calculators¶

Sleep calculators determine wait duration between agent wake cycles.

Example	Implementation
ai_based_sleep_calculator.py	LLM-driven calculation adapting sleep time based on conversation dynamics and engagement patterns.
static_sleep_calculator.py	Fixed-interval implementation providing consistent timing regardless of conversation state.
pattern_based_sleep_calculator.py	Keyword-driven timing adjustments based on conversation content and detected patterns.
function_based_sleep_calculator.py	Custom calculation logic implemented as a Python function with access to configuration and context.
custom_sleep_calculator.py	Full custom implementation extending the base `SleepTimeCalculator` class for advanced timing strategies.

Running Examples¶

Bash
git clone https://github.com/leomariga/ProactiveAgent.git
cd ProactiveAgent
uv sync

export OPENAI_API_KEY='your-api-key-here'
uv run python examples/minimal_chat.py

Contributing¶

Contributions are welcome. See CONTRIBUTING.md for guidelines.