Module summarizer 

LLM-based context summarization.

When the conversation grows beyond the context window, older messages are summarized into a compact representation to preserve important context while reducing token usage.

Structs

ContextSummarizer

Generates summaries of conversation history using the LLM.

ContextSummary

Summary of conversation context for compression.

TokenCostDisplay

Token and cost tracking display data.

Enums

SummarizeError

Errors during summarization.

TokenAlert

Token budget alerts.

Functions

smart_fallback_summary

Smart fallback summary that preserves structured information when LLM-based summarization fails. Instead of naive truncation, it extracts tool names, results, and preserves the first/last messages for continuity.