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. Each candidate is checked by a language model from its name, description and topics, and only actual MCP servers are listed (not clients, SDKs, lists or skills). Package metadata comes from ecosyste.ms (CC BY-SA 4.0). 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.
-
ai-cost-analyzerupdated this year
MCP server that helps developers understand and reduce their AI API costs through usage analysis, savings estimation, prompt optimization, and model comparison.
npm MIT -
FactAnchor-MCPgithub.com/Tausifonly001/FactAnchor-MCPupdated this year
Reduce AI hallucinations by 80% with zero-cost, local context anchoring (MCP server).
MIT ★ 1 -
Best Token-Efficient AI Coding Tools 2026 – Reduce Costs 90% with Local Memory & MCP
★ 1 -
shared-context-cache-mcp-servergithub.com/AiAgentKarl/shared-context-cache-mcp-serverupdated this year
MCP server for shared context caching — AI agents share computed results to reduce token cost and latency
pypi MIT ★ 1 -
Intelligent code knowledge graph for AI coding agents. Reduces cost by up to 44%, speeds up exploration by up to 57%, and boosts smaller model quality to match Opus.
npm MIT -
Bi-temporal, 4-tier hybrid cognitive memory engine for AI coding agents (Claude Code, Cursor, Windsurf) via FastMCP. Eliminates amnesia and reduces token costs using local PostgreSQL & Neo4j.
-
ollama-bridge-mcpgithub.com/ElnurIbrahimov/ollama-bridge-mcpupdated this year
MCP server that gives Claude Code 27 AI brains via Ollama — 15 cloud + 12 local models. Auto-routing, batch comparison, smart delegation. Reduce API costs, get specialized models for code, math, vision, reasoning.
MIT -
escalate_mcpgithub.com/EncryptShawn/escalate_mcpupdated this year
Reduce Coding costs, improve accuracy and performance. Have any AI agent escalate to Escalate_MCP where it will use skills, context7 data, the internet and powerful AI to analyze the deployment, update or bug and provide comprehensive guidance for a cheap AI to follow.
MIT -
gas-fakes-mcpgithub.com/tanaikech/gas-fakes-mcpupdated this year
This repository demonstrates how to dynamically add frequently used, AI-generated Google Apps Scripts to an MCP server as permanent tools. This approach enhances security, reduces costs, and improves efficiency for Google Workspace automation using the Gemini CLI and the gas-fakes library.
MIT ★ 11 -
fusion-mcp-hubgithub.com/KirtiJha/fusion-mcp-hubupdated this year
Solve the MCP token bloat problem. Build lean, custom MCP servers with only the tools you need from 300+ available tools. Reduce LLM costs and hallucinations. Includes CLI, WebSocket proxy, AI-powered recommendations, and interactive playground.
MIT ★ 2 -
pdf-decompiler-mcpgithub.com/noobieisgod/pdf-decompiler-mcpupdated this year
A MCP that turns PDFs into structured text, tables, figures, metadata, annotations, and page-level evidence for selective retrieval by MCP-compatible AI agents. Improves the information quality AI agents receive while also reducing token costs in some scenarios.
AGPL-3.0 ★ 17 -
vexp-cliactive
Local-first context engine for AI coding agents. Pre-indexes your codebase into a dependency graph and feeds any MCP agent only the code that matters, so it stops rediscovering your repo before every change. Fable-5-level results from Opus 5 on SWE-bench
npm UNLICENSED -
tokenwise-mcpupdated this year
Library + MCP server for reducing LLM token costs via compression, summarization, chunking, and Anthropic prompt-cache breakpoint planning
npm SEE LICENSE IN LICENSE -
awesome-confluence-mcpgithub.com/mazhar480/awesome-confluence-mcpupdated this year
🚀 Token-efficient MCP server for Confluence. Reduces LLM costs by 76% via Markdown conversion. Supports listing, searching, and fetching pages.
MIT ★ 3 -
toon-mcp-servergithub.com/elminson/toon-mcpupdated this year
MCP server for TOON (Token-Oriented Object Notation) - Reduce LLM token usage by 50-70%
npm MIT -
MCP bridge for AI agents using Entroly Context Assurance: reduce avoidable AI token usage with evidence selection, receipts, exact recovery, and verification.
npm Apache-2.0 -
omniagent-fleetgithub.com/landrover1984/omniagentupdated this year
AI Infrastructure OS — same engine, four facades (CLI, REST, web, MCP). Reduce 80-95% of AI spend by routing tasks to the right model.
pypi MIT -
Reduce Claude Desktop consumption by 10x - Integrate Google's Gemini or Z.ai's GLM-5 (744B params) with Claude via MCP for intelligent task delegation
MIT ★ 47 -
sym-protocolgithub.com/sym-protocol/sym-protocolupdated this year
SYM Protocol v1.0.1 — Symbolic prompt compression for LLMs. Reduce input tokens 40-65% without losing meaning. MCP server + CLI + Agent. Dynamic domain detection + history compression + stopwords.
npm MIT -
cctxgithub.com/anil7948/cctxupdated this year
Reduce Claude Code token usage by 70-90% using a free local LLM (Ollama). MCP server + Stop hook with codebase indexing, tool output compression, and turn summarization.
npm MIT ★ 3 -
An agentic code context engine MCP+CLI which pre-compiles an llm-described AST in the .tost format, greatly reducing context bloat and token costs for agentic coding tasks
MIT ★ 5 -
Suber-Agent-Teamgithub.com/stephenpham68/Suber-Agent-Teamupdated this year
MCP server that offloads Claude Code sub-agents to a cheap model (Haiku / any OpenAI- or Anthropic-compatible API) - run 20-100 parallel agents for fan-out, map-reduce & research without draining your Opus / Claude Max quota.
MIT ★ 3 -
tonl-mcp-bridgegithub.com/kryptomrx/tonl-mcp-bridgeupdated this year
Reduce LLM token costs by 30-60% with TONL format. TypeScript library & CLI with MCP Server for bidirectional JSON/YAML ↔ TONL conversion.
npm MIT -
🧠 Save 70–95% tokens in Claude Code, Cursor, Windsurf & Cline. Local Tree-sitter code graph MCP server: surgical edit context instead of full-file reads, refactor safety checks, and a hook that blocks large file reads. Reduce token usage & cost, zero LLM indexing.
pypi MIT ★ 4