Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.praison.ai/llms.txt

Use this file to discover all available pages before exploring further.

​
emergency_truncate

Method
This is a method of the ContextManager class in the manager module.
Emergency truncation when optimization isn’t enough. Aggressively removes messages to fit within target tokens. Preserves system messages and most recent turns.

​
Signature

def emergency_truncate(messages: List[Dict[str, Any]], target_tokens: int) -> List[Dict[str, Any]]

​
Parameters

​
messages
List
required
Messages to truncate
​
target_tokens
int
required
Target token count

​
Returns

​
Returns
List[Dict[str, Any]]
Truncated messages list

​
Uses

  • estimate_messages_tokens
  • estimate_message_tokens

​
Source

View on GitHub

praisonaiagents/context/manager.py at line 866