Crate bamboo_compression 

Token budget management for LLM conversations.

This module provides token budget management to prevent context window overflow while maintaining conversation continuity. It implements a hybrid approach combining rolling summary with recent message windowing.

Key Components

• types: Core types like TokenBudget, BudgetStrategy, PreparedContext
• counter: Token counting via BPE tokenizer (tiktoken) or heuristic estimation
• segmenter: Message segmentation preserving tool-call atomicity
• limits: Model context window limits registry
• preparation: Context preparation with budget enforcement
• summarizer: Conversation summarization for context preservation

Re-exports

pub use compression_tooling::active_messages_for_budget;

pub use compression_tooling::apply_compression_plan;

pub use compression_tooling::build_compression_plan_with_summary;

pub use compression_tooling::build_forced_compression_candidate_plan;

pub use compression_tooling::build_forced_compression_candidate_plan_with_fixed_tokens;

pub use compression_tooling::build_forced_compression_plan_with_summary;

pub use compression_tooling::build_message_summary_candidate_plan_with_fixed_tokens;

pub use compression_tooling::build_summary_prompt;

pub use compression_tooling::compression_summary_message;

pub use compression_tooling::context_window_usage_percent;

pub use compression_tooling::estimate_context_compression_exposure;

pub use compression_tooling::estimate_context_compression_exposure_with_fixed_tokens;

pub use compression_tooling::finalize_compression_candidate_plan;

pub use compression_tooling::normalized_summary_target_ratio;

pub use compression_tooling::normalized_trigger_percent;

pub use compression_tooling::summary_source_messages;

pub use compression_tooling::CompressionCandidatePlan;

pub use compression_tooling::CompressionPlan;

pub use compression_tooling::CompressionPlanError;

pub use compression_tooling::ContextCompressionExposure;

pub use compression_tooling::DEFAULT_SUMMARY_TARGET_RATIO;

pub use counter::HeuristicTokenCounter;

pub use counter::ProviderVisibleToolEstimate;

pub use counter::TiktokenTokenCounter;

pub use counter::TokenCounter;

pub use limits::create_budget_for_model;

pub use limits::ModelLimitsRegistry;

pub use preparation::estimate_prompt_cache_savings;

pub use preparation::estimate_prompt_cache_savings_with_fixed_tokens;

pub use preparation::prepare_hybrid_context;

pub use preparation::prepare_hybrid_context_with_fixed_tokens;

pub use retrieval_window::apply_retrieval_window_plan;

pub use retrieval_window::apply_retrieval_window_plan_with_trigger;

pub use retrieval_window::build_retrieval_window_candidate_plan;

pub use retrieval_window::build_retrieval_window_candidate_plan_with_fixed_tokens;

pub use retrieval_window::build_retrieval_window_candidate_plan_with_token_accounting;

pub use retrieval_window::build_retrieval_window_critical_overflow_plan_with_token_accounting;

pub use retrieval_window::effective_retrieval_window_target_tokens;

pub use retrieval_window::RetrievalWindowApplyError;

pub use retrieval_window::RetrievalWindowApplyResult;

pub use retrieval_window::RetrievalWindowCandidatePlan;

pub use retrieval_window::RetrievalWindowPlanError;

pub use retrieval_window::RetrievalWindowPolicy;

pub use retrieval_window::RetrievalWindowTokenAccounting;

pub use segmenter::MessageSegmenter;

pub use summarizer::HeuristicSummarizer;

pub use summarizer::Summarizer;

pub use summarizer::SummaryManager;

pub use summarizer::SummaryTrigger;

pub use types::BudgetError;

pub use types::PreparedContext;

Modules

compression_tooling

counter

Token counting for budget management.

limits

Model context window limits registry.

preparation

Context preparation for budget management.

retrieval_window

Pure candidate planning for retrieval-window context management.

segmenter

Message segmentation for budget management.

summarizer

Conversation summarization for rolling context management.

types

Core types for token budget management.

Structs

TokenBudget

Token budget configuration for a conversation.

TokenUsageBreakdown

Detailed token usage breakdown.

Enums

BudgetStrategy