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.
-
Stop your coding agent burning context on file reads. Delegates big reads to a cheap worker model and returns only the answer — 87-99% saved, measured per call. A hook enforces it in Claude Code; the MCP tools work in any client.
MIT -
Token-efficient MCP server for codebase architecture exploration. Save 70-90% tokens when an AI agent needs to understand a project.
npm MIT -
localthink-mcp-oldgithub.com/H3xabah/localthink-mcp-oldupdated this year
Local Ollama-backed MCP server for Claude Code. Query large files without burning context (~30x token savings)
MIT -
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 -
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 -
Token-saving layer on top of Graphify: symbol-level code slicing, context packs, a savings ledger with an honest baseline, hooks and MCP for Claude Code, Copilot and Codex.
pypi Apache-2.0 -
MCPMgithub.com/mikeychann-hash/MCPMupdated this year
MCP Memory reader that saves time and tokens that connects to LLMs and saves memories that the LLMs utilize to save time
-
MCP server for Angular: compiler-backed template answers, component contracts and version-accurate facts that save AI agents tokens
npm GPL-3.0 -
Local MCP server that saves tokens for AI coding agents. They query your codebase instead of reading whole files. Works with any MCP client.
Apache-2.0 ★ 1 -
Save 60-80% tokens when AI reads code — MCP server for token-efficient code navigation with AST-aware structural reading
npm ★ 5 -
@cocaxcode/token-optimizer-mcpgithub.com/cocaxcode/token-optimizer-mcpupdated this year
Orchestration + observability + coach layer for Claude Code token optimization. Measures tool usage, enforces budgets, advises on complementary tools (serena, RTK), and proactively surfaces savings tips.
npm MIT -
better-playwright-mcpgithub.com/ekuznetski/better-playwright-mcpupdated this year
Token-efficient Playwright MCP with getOutline and searchSnapshot - saves ~95% tokens vs full snapshots
-
mnemox-mcpgithub.com/ajaygh99/mnemox-extensionupdated this year
Mnemox MCP server — score prompts, count tokens, save/get memories. Works with Claude Code, Cowork, Cursor.
npm MIT -
Context optimization & token-saving MCP server for OpenCode CLI on free model tiers
MIT -
Thai-first context compression MCP server for Claude Code — 50-88% token savings, 100% recoverable
pypi MIT -
CLI and MCP server that verifies AI coding agent context-reduction proxies' token and cost savings claims
pypi Apache-2.0 -
Context-Canarygithub.com/amarisaster/Context-Canaryupdated this year
A token counter overlay for Claude. Warns you before context compaction hits so you can save what matters.
Apache-2.0 ★ 5 -
MCP server and hook toolkit that compresses tool output to save context window tokens.
npm MIT ★ 14 -
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 -
promptlint-mcpgithub.com/sean-sunagaku/promptlint-mcpupdated this year
Static linter for AI prompts — catch contradictions, strip fluff, save tokens. CLI + MCP server.
npm MIT ★ 1 -
json-to-toon-mcp-servergithub.com/your-username/json-to-toon-mcp-serverupdated this year
MCP server for converting JSON to TOON (Token-Oriented Object Notation) format - saves 30-60% on AI tokens
npm MIT -
cursor-translategithub.com/davlet42/cursor-translateupdated this year
Save tokens on Cyrillic text in Cursor: cheap nano translate tier, lazy EN doc cache via preToolUse hook, MCP, CLI
npm MIT -
Token-saving code editor for LLM agents: one MCP server, surgical range reads, atomic edits, syntax-gated saves - the same work for a fraction of the tokens.
MIT -
lean-readergithub.com/AIMento/lean-readerupdated this year
Turn any URL into clean, token-minimal text for LLMs — web tool + MCP server, with a per-call tokens-saved receipt.
npm MIT -
@iflow-mcp/glommer-cachebrogithub.com/glommer/cachebroupdated this year
File cache with diff tracking for AI coding agents. Drop-in replacement for file reads that saves ~26% tokens.
npm MIT -
Token-saving code search for AI coding agents — TypeScript, JavaScript & Python (zero setup), C#, C++, Go & 30+ more, via a language-server or tree-sitter index instead of grep. Local-only, no IDE. Claude Code plugin + MCP + CLI.
go MIT ★ 13 -
skimgithub.com/tonyzdev/skimupdated this year
Skim: A MCP server for Claude Code — skim large outputs, return only schema. Save context, save tokens.
★ 6 -
@molicherry/dokploy-mcpupdated this year
Dokploy MCP Server with Smart Skill routing - 265+ tools with intelligent categorization for 80-97% token savings
npm MIT -
claude-translategithub.com/davlet42/claude-translateupdated this year
Save tokens on Cyrillic text in Claude Code: cheap Haiku translate tier, lazy EN doc cache via PreToolUse hook, CLAUDE.md sync, MCP
npm MIT -
Number ontology with machine-checkable witnesses, plus AI software for development: exact data parser any_txt_parser and MCP gateway toolfmt (tool-output formatting, token savings)
★ 3 -
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 -
Token-saving browser automation MCP server: cached actions, then a local LLM, then Claude. Ground-truth benchmark included; 0 Claude tokens across 20 real runs.
-
toke-localgithub.com/rianprei/toke-localupdated this year
Zero-dependency MCP server that cuts round-trips and tokens for coding agents: 6 aggregate tools (agent_search, agent_read, agent_edit, agent_exec, agent_git, agent_savings)
MIT ★ 2 -
Daimonosgithub.com/beardfaceguy/daimonosno update data
Agent-optimized MCP server with structured JSON tools. 20-45% token savings.
mcpb -
Zero-cost token optimization engine for AI coding assistants (Claude Code, Cursor, Antigravity, Windsurf). Saves 70-95% tokens via AST skeletons, differential cache, and line slicing.
NOASSERTION ★ 6 -
Turn anything into token-optimized, layout-aware Markdown - and know exactly what it saved you. Lossless compression, honest token reports, CLI + MCP server + agent skill.
npm Apache-2.0 ★ 1 -
promptthrift-mcpgithub.com/woling-dev/promptthrift-mcpupdated this year
Smart token compression MCP server for LLM apps. Gemma 4 local compression, multi-model cost tracking, intelligent model routing. Save 70-90% on API costs.
MIT ★ 1 -
projectmindgithub.com/Nodemint-dev/projectmindupdated this year
Persistent, compact project memory for AI coding agents — MCP server + CLI. One command, works with Claude Code/Cursor/Windsurf/Gemini/Copilot, and shows you the tokens it saved.
npm MIT ★ 2 -
Audits AI agent logs for wasted tokens and returns a quantified savings report. Finds redundant calls, retry loops, and duplicate reasoning steps. Hosted MCP server.
MIT -
local-llm-delegation-mcpgithub.com/mrrodriguez/local-llm-delegation-mcpupdated this year
Intelligent task delegation for AI agents: Offload the "easy" work to your local machine and save premium tokens for the hard stuff.
★ 1 -
Token optimizer for Claude Code Gemini CLI & Qwen Code — compresses shell outputs via PreToolUse hook, tracks USD savings in TUI dashboard
MIT ★ 21 -
datoongithub.com/andrii-su/datoonno update data
Smart structured-data to TOON gateway: converts to TOON only when it saves LLM tokens.
pypi -
Token-saving Cyrillic markdown translation layer for Antigravity (agy): fast Gemini 3.7 Flash translate tier, lazy EN doc cache via PreToolUse hook, AGENTS.md / GEMINI.md sync, MCP and skills
npm MIT -
reddit-mcpgithub.com/Jing-yilin/reddit-mcpupdated this year
MCP server for Reddit API - Access Reddit data with TOON format for 90%+ token savings
npm MIT -
Hebbian memory for AI agents. Learns file patterns, predicts next tools, saves tokens.
npm -
tool-search-mcpgithub.com/ImBIOS/tool-search-mcpupdated this year
High-performance semantic tool search for Claude Code. Save 97% tokens with intelligent MCP tool discovery using embedding, BM25, and regex search methods.
★ 3 -
Self-documenting Model Context Protocol server for code context compression with real-time token savings statistics. Reduces LLM context by 70-98% while preserving semantic meaning. No system prompt configuration needed.
npm MIT -
code-contractor-mcpgithub.com/binacshera-ui/code-contractor-mcpupdated this year
🔧 Local MCP server for Cursor IDE - AST-powered code intelligence, 50%+ token savings, automatic backups
MIT -
SaveContextgithub.com/AlexanderBoger/SaveContextupdated this year
Git LFS for your LLM context — 20x Context Savings — an MCP server that keeps big documents out of the model and returns byte-exact quotes for a fraction of the tokens.
Apache-2.0 ★ 5 -
plane-mcp-servergithub.com/CaliLuke/plane-mcp-serverupdated this year
Plane's Unofficial MCP Server - streamlined responses that save tokens in your agent's context
MIT -
repo-context-mcpgithub.com/Jul879n/repo-context-mcpupdated this year
Universal MCP server that analyzes any codebase and provides structured context to AI assistants, saving tokens
npm MIT -
json-to-toon-mcp-servergithub.com/mhabedini/json-to-toon-mcp-serverupdated this year
MCP server for converting JSON to TOON format - saves 30-60% on AI tokens
MIT ★ 1 -
Graph-backed session memory for LLMs: checkpoint, resume, file analysis, token-savings stats.
pypi -
Token-efficient code generation pipeline - parallel implementation with hosted LLM (Cerebras) for ~60% token savings. Includes MCP server.
MIT ★ 3 -
Engramgithub.com/ckwich/Engramupdated this year
MCP memory OS with deep graphing, document digestion and token-saving power. Context limits are no longer an issue.
MIT -
MCP server with a context ledger: cryptographically signed receipts of exactly which file versions the model was shown. Honest benchmarks: no real-world token savings (-1.4% measured); ~72% applies only to a synthetic re-read loop.
npm MIT ★ 4 -
MCP plugin focused on boosting agent context efficiency and saving tokens
MIT ★ 1 -
snapshot-mcp-servergithub.com/WhenMoon-afk/snapshot-mcp-serverupdated this year
Save and resume MCP conversations with token-efficient local snapshots.
npm MIT -
Token-efficient MCP server for tabular data retrieval. Index CSV/Excel files, query rows, aggregate — 99%+ token savings vs raw file reads.
pypi NOASSERTION ★ 83 -
Autonomous multi-agent MCP orchestration platform powered by LangGraph state machines, 8 specialist ninja agents, SQLite FTS5 skill indexing, and per-turn token governors saving 83–98% context across 8 AI coding clients
npm Apache-2.0