Find the ghost tokens. Fix them. Survive compaction. Avoid context quality decay.
-
Updated
Nov 16, 2026 - Python
Find the ghost tokens. Fix them. Survive compaction. Avoid context quality decay.
Hook-based token compressor for 7 AI CLI hosts (Claude Code, Copilot CLI, OpenCode, Gemini CLI, Codex CLI, Pi, Hermes). Up to 95% bash compression, signature-mode for code reads, cross-call dedup, MCP server, self-teaching protocol. Zero runtime deps.
TOON encoding for Laravel. Encode data for AI/LLMs with ~50% fewer tokens than JSON.
token-ninja routes deterministic shell commands locally — zero LLM calls, ~19µs latency. Works silently inside AI tools via MCP.
Deterministic AST Context Compressor & MCP Server for Cursor, Claude & Local LLMs. Cuts prompt tokens by 85–95% with 100% Zero-Leak security.
Compact coding-agent context without summarizing it away. Verbatim, offline, with a receipt and recall for every cut. CLI, MCP server, Claude Code plugin. MIT.
Compress logs for LLM analysis — Rust-powered, Python API. 40-60% token savings.
Cut your Claude Code token bill: tool-output trimming, read dedup, model-tier discipline, and live cost + cache-hit stats.
Maximum Claude output with minimum tokens. 50+ compression patterns, 10 system prompts, benchmarked results. Star to unlock usage.
A modular toolkit for Claude Code to optimize context, planning, and agent execution with lower token usage and less repeated codebase discovery.
astro-agent-optimized(AAO) is agent-optimized output for Astro JS tools.
A simple open source vibecoded prompt optimizer for personal use, you are free to use it as needed.
⚡ Universal IDE plugin & proxy that cuts LLM token consumption by up to 70% using deterministic AST pruning & context compression. Written in Go.
A skill that makes LLMs think faster and cheaper: up to 77% fewer thinking tokens and answers up to 2x faster, with the same quality. Benchmarked on Claude, Grok, Gemma, Qwen and gpt-oss.
Trim Python code for LLMs by removing comments, docstrings, and whitespace. Save AI tokens, reduce API costs, and fit more code into ChatGPT, Claude, and Gemini.
⚡ The Featherweight AI Browser Engine in Rust & Python — Dual-engine web browsing for AI agents & LLMs with 90% fewer tokens and native MCP support.
High-performance HTTP reverse proxy, secret redaction firewall, and two-tier caching engine for AI coding agents and LLM applications.
Smartest context aware Token Optimizer for mcp tools.
⚡ The ultimate Grok & Grok Bot CLI: multi-account parallel orchestration, automatic quota failover, 3-layer export, and local OpenAI proxy.
Auto-route Claude Code prompts to Haiku, Sonnet or Opus. Learns from escalations, saves tokens.
To associate your repository with the token-optimizer topic, visit your repo's landing page and select "manage topics."