MCP servers that ship as code — a repository or a package you install and run locally (usually over stdio). They have no hosted endpoint, so we cannot probe or grade them: instead each one shows its repository, install command, license, stars and when it was last updated.
Collected from GitHub repositories and npm, PyPI, Cargo, Go, Maven, NuGet and RubyGems packages tagged as MCP servers, the official MCP Registry, Docker's MCP catalog, ModelScope, Baidu MCP World, Cline and other public directories; one entry per repository. A server that also offers a hosted endpoint is listed under Remote instead. These entries are not counted in our reachability, verification or readiness statistics.
-
Cuts Claude Code and LLM token usage by 60-85% via AST navigation, deduplication, and context-budget optimization.
-
Zero-dependency context & token optimizer for Claude Code, Codex CLI, and Google Antigravity. Cuts terminal output by up to 96.4% to prevent context window exhaustion.
npm MIT -
adsagent-mcpgithub.com/nowork-studio/adsagent-mcpupdated this year
Connect your Google Ads to Claude, Codex, or any MCP-compatible AI assistant. Manage campaigns, optimize keywords, and cut wasted spend through conversation.
MIT -
Apple Ads for AI agents: audit campaigns, cut taps that never install and follow a ranked optimization plan. Plugin and remote MCP server for Claude, ChatGPT, Codex, Cursor, Gemini CLI and Grok.
MIT -
NestSolver's hosted MCP server: cut lists for bars and sheets, true-shape nesting, and pallet and truck loads, from any MCP client.
MIT -
Cabinet furniture design: parts, sheet cutting, drilling, drawings. Opens and exports Bazis-Mebelshchik models. Desktop app, CLI and MCP server for AI. Early alpha.
MIT ★ 7 -
Free, open-source 2D nesting for fixed sheets: laser, CNC, plasma, waterjet and fabric. SVG/DXF in and out, part-in-part, several sheets. Runs in your browser (nothing uploaded), on the command line and as an MCP server. Built on sparrow and jagua-rs.
MIT ★ 17 -
Расчётный движок каркасного дома для ИИ-агентов: MCP-сервер, физические доски, нагрузки, раскрой 6 м, смета, 3D-просмотрщик и альбом КР
-
token_optimizergithub.com/omkar9854/token_optimizerupdated this year
Zero-dependency, fully-reversible context compression for AI agents — cut LLM token usage up to 94% (JSON tabularization, log template mining, query-aware code skeletons). Stdlib-only, air-gap ready, built-in MCP server.
Apache-2.0 ★ 1 -
Portable MCP server: weave LLM context without cutting a thread.
pypi MIT -
Auragithub.com/IzzoSol/Auraactive
Cut LLM token spend with zero dependencies. AURA answers recurring prompts free from cache, compresses conversation history, and distills bloated system prompts. CLI, MCP server (Claude, Cursor, Claude Code) and library.
MIT ★ 3 -
tokensavergithub.com/pozii/tokensaverupdated this year
MCP server that cuts AI agent token costs by up to 97% — compression, caching, pruning, web extraction
Apache-2.0 ★ 1 -
Local MCP code search that cuts AI agent tokens: AST chunks, CPU reranking and a dependency impact graph. -63.9% tokens measured.
MIT ★ 1 -
@kunal_dutt/token-lensactive
Count and cut the tokens your prompts, chats and agents use, adapted to the model and task. MCP server and self-hosted gateway. Collects no data.
npm MIT -
Agent Intermediate Language (AILv1): Zero-loss AI context compression protocol, AST signature extractor, and MCP server that cuts LLM handoff tokens by 85–95%.
MIT ★ 1 -
Claude Code plugin for MinniMemoryMCP (npm: minnimemory), which cuts Claude Code and Codex token usage. Generated from release builds.
npm NOASSERTION -
High-performance code-intelligence engine for AI agents and IDE, supports 257 languages, multi repositories, based on graph, with access via CLI, MCP Server, and API. AI coding agents teammate - expose only needed information, cutting token usage up to 50x. 100% local. Discord: https://discord.gg/39MFHu3J5d
npm Apache-2.0 ★ 2,077 -
Cut the noise from command output before it reaches your AI agent. Errors always kept, full output one call away. MCP server + CLI for any agent. MIT.
MIT -
Universal LLM router for AI coding tools. Works with Claude Code, Cursor, Codex, Gemini CLI, Copilot and more. Free-first fallback chain cuts costs 35-80%.
pypi MIT ★ 129 -
Local Tree-sitter code knowledge graph & MCP server that gives Claude Code, Cursor and other AI coding agents surgical edit context — cutting token usage by 80-95%.
pypi MIT -
Request-path token proxy + analyzer for agent conversations. Live: cut 74% of input tokens on a real DeepSeek call (mcp-token-saver-proxy.fly.dev). MIT core, no-LLM, crypto-native Pro.
-
Cut LLM agent costs 40-70%: route easy work to free models, keep the paid model for decisions and review. Research, playbook and a zero-dependency MCP cascade server for Codex/OpenCode.
MIT ★ 1 -
💎 Semantic Context Compressor, Live Token/USD Cost Calculator & Adaptive Model-Tier Router MCP Server — Cut LLM token bills by 85% across Claude Code, Cursor, VS Code, Codex & Antigravity.
MIT ★ 1 -
Drop-in OpenAI / Anthropic / Gemini BaseURL Proxy & Edge MCP Server (~0.04ms). Cut LLM input tokens by 60–99% via AST code skeletons, JSON pruning & stream healing with 1 line of config.
MIT ★ 1 -
A decoupled local-SLM pre-processing and routing layer for MCP and cloud LLMs. Intercepts, distills, and routes prompts locally with Ollama to cut token costs and protect monthly quota.
npm MIT -
Cut Claude Code costs: Claude stays the brain while its subagents run on cheaper models (OpenRouter, DeepSeek, Qwen, GLM, Kimi). Rust CLI, local proxy and MCP server with secret/PII redaction, budgets and savings reports.
MIT ★ 1 -
See what your MCP servers cost you in tokens — and cut it. Records every MCP tool call, itemises the per-session tax each server charges, and tells your coding agent what to remove. All local, zero config.
npm MIT