CICADA is an MCP server that provides structured code indexing for AI code assistants. Through AST-level indexing, call site tracking, and semantic search, it provides efficient context compression for Elixir, Python, and Erlang code repositories, reducing token usage and improving the quality of code understanding.

An optimized MCP protocol web scraping server that significantly reduces the token usage of AI processing through server - side HTML to Markdown conversion and CSS selector filtering, providing efficient content extraction functions.

An MCP server for Claude Code that achieves over 90% context compression through structured incremental snapshots, significantly saving token usage.

The MCP Chain of Draft Prompt Tool is a tool that enhances the reasoning ability of LLMs. By converting standard prompts into Chain of Draft or Chain of Thought format, it significantly improves reasoning quality and reduces token usage.

An MCP server implementing the Chain of Draft inference method, significantly reducing token usage while maintaining accuracy

TokenScope is a token-aware directory exploration tool for large language models, providing intelligent directory structure analysis, token-aware file content extraction, token usage statistics, and comprehensive report generation functions.

The TOON MCP server is a model context protocol server for TOON encoding. It converts structured data into a token - optimized representation through a compact data format, which can reduce LLM token usage by 50 - 70%, thereby significantly reducing AI call costs.

Token Optimizer MCP is an intelligent token optimization MCP server that can reduce context window usage by 60-90% through intelligent caching, compression, and tool replacement. It supports AI tools such as Claude Code and Claude Desktop.

The Notion MCP server is a middleware service that enables interaction between LLMs and the Notion workspace through the Notion API, supporting Markdown conversion to optimize token usage efficiency.

An MCP server based on the cursor-agent CLI, providing tools for code analysis, search, planning, and editing to help Claude Code reduce token usage and costs

An MCP server that optimizes token usage in Claude Code by automatically compressing JSON, Markdown, and code files, reducing token consumption when reading files.

The ClickUp MCP server focuses on task management. It provides API response optimization through 37 tools, automatically streamlining verbose JSON data to core content to save token usage.

A local MCP server that specifically provides DaisyUI component documentation, enabling efficient token usage through tool calls and supporting queries for component lists and access to detailed documentation.

An MCP server specifically designed to optimize token usage of AI agents, improving efficiency by converting data to the TOON format and removing non-essential content from code files