Context Window 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.

jcodemunch-mcp

Cut AI token costs 95%+ on code exploration. The leading MCP server for precise, symbol-level GitHub code retrieval via tree-sitter AST. Works with Claude Code, Cursor & any MCP client. 313B+ tokens saved.

cocoindex-code

A super light-weight embedded code search engine CLI (AST based) that just works - saves 70% token and improves speed for coding agent 🌟 Star if you like it!

token-goat

Token burn reducer and focus keeper for Claude Code, Codex, Gemini CLI, Cline, Windsurf, Aider, Cursor, Copilot, and more: surgical read hints, PDF/Office/CSV/markdown file interception, 160+ filter & interception rules, compact manifest injection, image shrinking, cache and compact skills, cache MCP calls, and much more.

toolrecall

High-performance local SQLite FTS5-based tool output cache and MCP multiplexer for AI agents.

shardstitch

Recover and hand off your AI coding session — after a rate limit, crash, or context bloat — across 23 tools. 100% local, no cloud, no account.

tokenpak

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

diffctx

Smart git diff context for LLMs - selects the minimal code fragments needed to understand a change. Also exports full codebase in YAML/JSON/MD.

zing-audit

LLM relay reality check — a CLI that audits whether an API relay actually serves the model it claims (货不对板检测)。

infiniloom

Python bindings for Infiniloom - AST-aware code context engine for RAG and vector databases. Tree-sitter parsing, PageRank ranking, BLAKE3 content-addressable chunks, 27+ tokenizers.

pagetomd

Convert web pages to clean Markdown

mindcore-memory

Production-hardened MCP memory server for AI agents: hybrid BM25+FAISS search, circuit breaker, SLO tracking, Prometheus metrics, BND boundary evaluation. Gives AI persistent long-term memory that survives across sessions.

treemapper

TreeMapper product layer on top of the diffctx engine

kopipasta

`cat project | LLM | patch`. Transparent context control and interactive patching for terminal-centric LLM workflows.

contextrot

Find out where your coding agent starts degrading. Personal context-rot analytics from your own sessions - 100% local, zero-config: uvx contextrot

a2cr-mcp

MCP server for client-encrypted AI agent handoffs with WorkBaton and WorkStash

mnemara

Controlled rolling-context conversation runtime for Claude. Stop your agent from running away with your bill.

copium-ai

Context optimization layer for LLMs. 65-90% token savings with zero quality loss. Drop-in proxy for Claude, GPT, Gemini, and local LLMs (Ollama/VLLM/llama.cpp). Features KV cache-aware compression, session dedup, error cards, and Pichay-proven context paging.

headroom-hardened

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

ultracontext

Open Source Context infrastructure for AI agents. Auto-capture and share your agents' context everywhere.

codec-cortex

Deterministic structural compression protocol for LLM agent cognitive memory. 85% less tokens. Zero LLM in codec cycle. 100% reversible. Golden ratio memory distribution.

lethe-llm-context

Context garbage collector for long-running LLM agents - offload & recall to save tokens in Claude Code and Codex via MCP.

contextzip

CLI tool for intelligent project packaging with framework detection, smart exclusions, and git-aware file selection.

rm-contextos

Open-source context operating system for AI coding agents. Scans repositories, builds repo intelligence, selects task-specific context under token budgets, and exports optimized context packs for Claude Code, Codex, Cursor, Aider, and other coding agents.