bamboo-compression 2026.9.15

Compression utilities for Bamboo sessions and memory workflows
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45

//! 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

pub mod compression_tooling;
pub mod counter;
pub mod limits;
pub mod preparation;
pub mod segmenter;
pub mod summarizer;
pub mod types;

pub use compression_tooling::{
    active_messages_for_budget, apply_compression_plan, build_compression_plan_with_summary,
    build_forced_compression_candidate_plan,
    build_forced_compression_candidate_plan_with_fixed_tokens,
    build_forced_compression_plan_with_summary, build_summary_prompt, compression_summary_message,
    context_window_usage_percent, estimate_context_compression_exposure,
    estimate_context_compression_exposure_with_fixed_tokens, finalize_compression_candidate_plan,
    normalized_summary_target_ratio, normalized_trigger_percent, summary_source_messages,
    CompressionCandidatePlan, CompressionPlan, CompressionPlanError, ContextCompressionExposure,
    DEFAULT_SUMMARY_TARGET_RATIO,
};
pub use counter::{
    HeuristicTokenCounter, ProviderVisibleToolEstimate, TiktokenTokenCounter, TokenCounter,
};
pub use limits::{create_budget_for_model, ModelLimitsRegistry};
pub use preparation::{
    estimate_prompt_cache_savings, estimate_prompt_cache_savings_with_fixed_tokens,
    prepare_hybrid_context, prepare_hybrid_context_with_fixed_tokens,
};
pub use segmenter::MessageSegmenter;
pub use summarizer::{HeuristicSummarizer, Summarizer, SummaryManager, SummaryTrigger};
pub use types::{BudgetError, BudgetStrategy, PreparedContext, TokenBudget, TokenUsageBreakdown};