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.
-
claude-token-saver-mcpgithub.com/BlackFoil/claude-token-saver-mcpupdated this year
MCP server that offloads coding tasks to local LLM (Ollama) to save Claude API tokens
npm Apache-2.0 ★ 3 -
@daniel.tinois/token-saveractive
MCP que adapta contexto e estratégia à complexidade de tarefas de desenvolvimento.
npm ISC -
Minimal read-only MCP data bridge for ChatGPT Pro planning and Codex verification/implementation, with benchmark and acceptance tests.
-
token-saver-mcpgithub.com/flightlesstux/token-saverupdated this year
MCP plugin that fires warnings, errors, and alerts when AI token usage is wasteful. Works with any MCP client. Auto-suppresses logs and ignored history to keep context lean.
npm MIT -
Reduce Claude Code and OpenAI Codex token usage with deterministic, recoverable tool-output compaction and MCP tools.
MIT -
MCP server for Angular: compiler-backed template answers, component contracts and version-accurate facts that save AI agents tokens
npm GPL-3.0 -
mcp-token-saverupdated this year
Local MCP server that surfaces real-time Claude.ai subscription usage (5h/7d), forecasts burn rate, and gates expensive operations before they run.
npm MIT -
local-token-savergithub.com/itsarmaanverma/local-token-saverupdated this year
Locally installed, folder-scoped retrieval-first context for Claude Code / Codex — SQLite FTS5+vector hybrid search, PDF→Markdown→vectorizer pipeline, MCP server.
MIT ★ 2 -
@csc-protocol/token-saveractive
Autonomous AI agent tool and MCP server for semantic token compression and L402 edge solvency clearance.
npm UNLICENSED -
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.
-
Real-time Claude.ai/Codex subscription awareness for AI coding assistants. Surfaces your live Session 5hr and Weekly 7day utilization, forecasts when you'll hit the limit, gates expensive operations before they run, and measures real per-task cost — all without leaving your machine.
MIT ★ 6 -
Curated, measurement-driven collection of token-saving MCPs for Claude Code, Codex, Cursor & Windsurf. Each MCP shipped here has passed Humanswith.ai's deterministic-compressor benchmark with documented byte-saving + quality-preservation measurements.
MIT ★ 1 -
Session memory & context enhancer for chat.z.ai — MCP server (12 tools), local vector memory, token saver, smart skills, GitHub private-repo cloud DB. Built with GLM 5.3 Flash.
MIT -
Universal token saver for AI coding agents: compresses every tool output (shell, MCP, Roblox Studio, grep, big files) before the model reads it. One Rust binary.
MIT -
Remote MCP server run by an AI bot: free token-saver tools + pay-per-call x402 tools
MIT -
quotarelaygithub.com/quotarelay/quotarelayupdated this year
Local MCP context engine for coding agents: bounded repo context, memory, cache proof, handoff packets, and token-saver demos without provider calls.
Apache-2.0 -
graphsiftgithub.com/maheshmakvana/graphsiftupdated this year
Token Saver for Claude, GPT-5 & Gemini. 80-150x code context reduction, F1 0.85. AST dependency graph, ranked context selection, 19 CLI compressors, MCP server, agent memory. Save LLM tokens — zero telemetry.
pypi MIT ★ 6 -
context-enginegithub.com/JaimeJunr/context-engineupdated this year
Drop-in token saver for Claude Code & MCP agents — auto-rewrite hooks + 10-tool MCP server + BM25/PageRank repo maps + call graphs in 7 languages + local RAG. Single Rust binary, 100% local, zero APIs.
Apache-2.0 ★ 2 -
claude-context-savergithub.com/dongnh311/claude-context-saverupdated this year
MCP server that cuts Claude Code token usage: log compression (builds/tests/installs) + smart file reading (symbol-level source access via tree-sitter).
npm MIT ★ 2 -
The Context-Saver: Multi-Server Multiplexer (Aggregator) MCP Server. One gateway connection instead of dozens — lazy tool schemas cut MCP context burn by ~90%.
npm MIT -
cheap-llm-mcpgithub.com/stblackcat/cheap-llm-mcpupdated this year
Save premium-model tokens by routing simple tasks to cheap OpenAI-compatible LLMs.
npm MIT -
reposynapsegithub.com/jul879n/reposynapseupdated this year
Universal MCP server that analyzes any codebase and provides structured context to AI assistants, saving tokens by eliminating redundant project exploration
npm MIT -
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 -
reporosegithub.com/mibrahimpro/reporoseupdated this year
Repository scanner that builds a structural map of a codebase (functions, imports, exports, dependencies) and adds file descriptions of what the file does,, so anagentic AI can just read the map and know the project context
npm GPL-3.0-or-later -
oioxo-mcpgithub.com/aradsoftca/oioxo-mcpupdated this year
OIOXO — cut your AI bill by up to ~90%: it feeds Claude Code, Copilot, Cursor and any MCP agent only the relevant slice of your codebase instead of whole files. Also a private terminal coding agent (`oioxo code`, on-device with --local or any key). Runs o
npm SEE LICENSE IN LICENSE.md