Token Optimization Python Packages

headroom-ai

Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 60-95% fewer tokens, same answers. Library, proxy, MCP server.

tokenpak

Drop-in HTTP proxy that compresses LLM context, optimizes cache hits, routes smart, and tracks every dollar. Zero SDK changes required.

entroly

Open-source context engine that catches AI hallucinations and cuts your token bill 70–95%. The only AI helper that shows its work. Claude · Cursor · Codex,GPT & Custom Providers

entroly-core

Open-source context engine that catches AI hallucinations and cuts your token bill 70–95%. The only AI helper that shows its work. Claude · Cursor · Codex,GPT & Custom Providers

agent-friend

Find the MCP schema issues eating your context window. 156 checks. Grade A+ through F. ESLint for MCP.

rtk-hermes

RTK plugin for Hermes — rewrites shell commands for 60-90% LLM token savings

precis-mcp

MCP server giving LLM agents a seven-verb API over papers, documents, code, state, patents, and cached web/Wolfram/YouTube tool calls

anvl-monitor

Session monitor for Claude Code — detects inflated sessions, saves quota with smart rotation and handoffs. Growth-aware health tracking with dual-signal waste detection.

tok-protocol

tok is an invisible bridge

context-router-cli

Local-first CLI and MCP server for AI coding context — minimum useful context, maximum agent performance

octave-mcp

OCTAVE protocol - structured AI communication with 3-20x token reduction. MCP server with lenient-to-canonical pipeline and schema validation.

graph-tool-call

Graph-based tool retrieval for LLM agents — 248 tools → 82% accuracy, 79% fewer tokens. Zero dependencies. OpenAPI / MCP / LangChain.

sqz

Compress LLM context to save tokens and reduce costs

tailspin-ai

Token optimization and compression for Claude API requests

agentatlas

Shared browser interaction schema registry for AI agents. 80-100% token reduction.

mnemon-ai

Cut LLM agent token costs by 93%. Execution cache for LangChain, CrewAI, AutoGen — 2.66ms vs 20 seconds, zero tokens on repeat runs.

ctxeng

Build perfect LLM context from your Python codebase — automatically

prism-cc

Session intelligence for Claude Code - find extra token usage, why your sessions fail, and how to fix it.

llm-prompt-refiner

🚀 Lightweight Python library for building production LLM applications with smart context management and automatic token optimization. Save 10-20% on API costs while fitting RAG docs, chat history, and prompts into your token budget.

vecr-compress

Deterministic, auditable LLM context compression — regex whitelist guarantees structured facts (IDs, URLs, dates, code) survive. Two layers: retention + heuristic knapsack.

entroplain

Entropy-based early exit for efficient agent reasoning. Stop burning tokens. Know when your agent has finished thinking.

llama-index-postprocessor-vecr

Deterministic, auditable LLM context compression — regex whitelist guarantees structured facts (IDs, URLs, dates, code) survive. Two layers: retention + heuristic knapsack.

lattice-transport

A complete transport layer for LLM traffic

mnemosyne-engine

State aware knowledge compression, ingestion, and hybrid retrieval engine. Zero dependencies. Sub-100ms queries.