> ## Documentation Index > Fetch the complete documentation index at: https://praison.ai/docs/llms.txt > Use this file to discover all available pages before exploring further. # Context Manager • AI Agent SDK > ContextManager: Unified facade for context management. # ContextManager > Defined in the [**manager**](../modules/manager) module. AI Agent Unified facade for context management. Orchestrates budgeting, composition, optimization, and monitoring. Provides hooks for exact LLM boundary snapshots. Tracks optimization history for debugging. ## Constructor No description available. No description available. No description available. No description available. No description available. No description available. ## Methods Process messages through the context pipeline. Capture exact state at LLM call boundary. Register a callback for LLM boundary snapshots. Get the last LLM boundary snapshot. Estimate tokens with optional validation. Get token budget for a specific tool. Set token budget for a specific tool. Truncate tool output according to its budget. Get optimization history. Get current context statistics. Emergency truncation when optimization isn't enough. Get the fully resolved configuration with source info. Reset manager state. ## Usage ```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} manager = ContextManager(model="gpt-4o") # Process messages before LLM call result = manager.process( messages=messages, system_prompt=system_prompt, tools=tools, ) # Get optimized messages optimized_messages = result["messages"] # Check if optimization occurred if result["optimized"]: print(f"Saved {result['tokens_saved']} tokens") ``` ## Source `praisonaiagents/context/manager.py` at line 292 *** ## Related Documentation