TokenZero MCP Server

by tokenzero Your server? Claim it

answering

TokenZero is answering right now. Last checked 14 min ago. 3 345 installs a week from npm. It exposes 6 tools. Last commit 1 Nov 2026.

LLM cost optimization gateway, repo token-waste auditor & model arbitrage MCP server.

Installs per day peak 1 669 · avg 111

a month agotoday

Uptime history 7 days of history

7 days agonow

100.0%

Uptime 24h

91 of 91 checks

Tools

read from the server

814 ms

Response time

average over 24h

3 345

Installs / week

npm and PyPI

What changed 12

Every tool that appeared, vanished or quietly changed what it asks for. Recorded since 27 October 2026. No other catalogue keeps this.

2 Nov a tool changed version

1 Nov a tool changed version3 times that day

1 Nov a tool appeared try_tokenzero

1 Nov a tool description was rewritten calculate_model_arbitrage

28 Oct 3 tool descriptions were rewritten audit_repo_token_waste, calculate_model_arbitrage, optimize_prompt_payload

28 Oct a tool changed the parameters it asks for calculate_model_arbitrage

28 Oct a tool changed version

27 Oct a tool changed version

and 2 more, back to 27 October 2026

What the code does

We read the source, 21 h ago · tools taken from the live server · rules 995398eba69f

Capabilities

What this server is able to do. For an MCP server this is often the job itself — a terminal server runs commands because that is what it is for. Listed so you know what you are plugging in, not as an accusation.

Reads files and sends them to the network mcp/bin/tokenzero-mcp.mjs:904, [пакет] bin/tokenzero-mcp.mjs:904

      entries = fs.readdirSync(dir, { withFileTypes: true });

Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.

This code can reach further than it looks

We found places where it runs commands, builds paths or queries from values it is given. None of that is a flaw by itself — it becomes one when the code changes, and code changes quietly between releases. We re-read it on every one.

Connect this server

Endpoint below is the one we actually reach during checks — not the one copied from a README. Last verified 14 min ago.

Claude Code Claude Desktop Codex Cursor VS Code

run in your terminal

claude mcp add mcp --transport http https://api.tokenzero.sh/mcp

Settings → Connectors → Add custom connector, then paste this address

https://api.tokenzero.sh/mcp

~/.codex/config.toml

[mcp_servers.mcp]
url = "https://api.tokenzero.sh/mcp"

.cursor/mcp.json

{
  "mcpServers": {
    "mcp": {
      "url": "https://api.tokenzero.sh/mcp"
    }
  }
}

.vscode/mcp.json

{
  "servers": {
    "mcp": {
      "type": "http",
      "url": "https://api.tokenzero.sh/mcp"
    }
  }
}

Available tools 6

Read directly from the server with tools/list, grouped by what they act on. If a tool disappears, we record the date.

audit

audit_repo_token_waste

No API key needed. Static-analyze a public GitHub repo (e.g. {"repo": "owner/name"}) or a set of source files for LLM token waste: uncached static prompts, frontier models on extraction tasks, hot loops, non-deterministic repeat calls, and prompt-caching mis-integration (caching gated on model names, volatile values in the system prompt, random ids in history, wrong-provider cache markers, misplaced breakpoints, cost math that ignores cached tokens). Returns findings with fixes and an estimated recoverable percentage.

calculate

calculate_model_arbitrage

No API key needed; only `model` is required (e.g. {"model": "gpt-6.1-sol"}, defaults 2,000 tokens in and 400 out). Price a workload at list prices across models: cost per request and per month for the requested model, its cheaper sibling, and the effect of prompt caching. The result names the pricing table version it used.

optimize

optimize_prompt_payload

No API key needed. Rewrite an OpenAI Chat/Responses or Anthropic Messages request body so its static prefix is provider-cacheable (Anthropic cache_control breakpoints, OpenAI prompt_cache_key). Returns the optimized body, what changed, the token profile and the projected per-request saving. Never changes the tokens the model sees.

provision

provision_workspace

Create a TokenZero workspace and key with no form. Returns the key once, the base URLs that route through it (OPENAI_BASE_URL / ANTHROPIC_BASE_URL), the one-click dashboard link and the badge URL. Free up to the monthly free tier; nothing is billed.

try

try_tokenzero

Call this first: no API key, no signup, no setup. Shows what TokenZero would do to one of your LLM calls: it runs the four savings engines (prompt-cache formatting, edge cache, cascade advice, signed receipt) on a sample or your own prompt and model, and returns the list-price saving, the signed receipt and a copy-paste keyless curl that makes a real call on a free model. Pass {} for the default, or {"model": "gpt-6.1-sol", "system": "...", "prompt": "..."}.

workspace

workspace_metrics

Live value scorecard for a TokenZero workspace: hard savings and net ROI per window, the four engines' contribution, speed, parity, availability and privacy posture, per-model breakdown and one-click recommendations. Needs the workspace key (tz_live_…); pass demo: true for a labelled sample.

Endpoints

URL	Transport	State	Latency	Checked
https://api.tokenzero.sh/mcp	streamable-http	answering	803 ms	14 min ago

Alternatives to TokenZero

same job, measured the same way

TokenZero

by tokenzero-ai

LLM cost optimization gateway, repo token-waste auditor & model arbitrage MCP server.

6 tools answering

Skim MCP Server

by luw2007

MCP server for Skim code transformation. Compresses code 60-95% for LLM context optimization.

54 installs/wk local only

Github MCP Server

by crypto-ninja

AI-optimized GitHub MCP server: 112 tools, 98% token reduction, compact responses.

270 installs/wk local only

Azure FinOps MCP Server

by raviteja-pegata

Azure FinOps MCP: cost, budgets, forecasting and optimization for Claude, VS Code and Cursor.

98 installs/wk local only

Multi-MCP

by religa

Multi-model AI orchestration MCP server with code review, compare, and debate tools.

99 installs/wk local only

Arm MCP Server

by arm

Official Arm MCP server for code migration, optimization, and Arm architecture guidance

local only

Figma To Code MCP

by felixanhalt

LLM-optimized Figma MCP server.

836 installs/wk local only

Labby

by dinglebear

Rust MCP gateway with Code Mode, authentication, setup, logs, CLI, HTTP API, and operator web UI.

303 installs/wk local only

TokenZero — questions

Answers built from our own checks of this server.

What can TokenZero do?

It exposes 6 tools, read directly from the server on our last check. Among them: audit_repo_token_waste, calculate_model_arbitrage, optimize_prompt_payload, provision_workspace, try_tokenzero, workspace_metrics. The full list with descriptions is on this page — we take it from the server itself via tools/list, not from a README. How MCP servers expose tools in the first place →

Is TokenZero working right now?

We send a real MCP handshake every 15 minutes. Over the last 24 hours 91 of 91 checks got a reply (100.0%), average response time 814 ms. The bar chart above shows every period we have measured.

How do I connect TokenZero?

Copy the ready config from this page. We generate it for Claude Code, Claude Desktop, Codex, Cursor and VS Code, each in the form that client actually reads. It is a remote server, so there is nothing to install — the client connects to the address.

Does TokenZero need an API key?

No. TokenZero completed a full MCP handshake with us as an anonymous client and listed its tools without asking for anything. All 6 of them are readable on this page. This is what we observed, not what the docs claim.

How fast is TokenZero?

It answers our handshake in 814 ms on average, which is faster than 10% of all working MCP servers we measure. That is on the slow side — worth knowing if the tool sits inside an interactive loop. The comparison comes from our own checks across the whole registry, every 15 minutes.

How many people use TokenZero?

The npm package @tokenzero/mcp was installed 3 345 times in the last week. We show installs rather than GitHub stars on purpose: a star is a bookmark, an install is someone actually running it.

Is TokenZero open source?

The code is open source: it is published under the Apache-2.0 licence, written in JavaScript, 0 stars on GitHub and 1 open issue. The licence covers the code in the repository, not the hosted service at api.tokenzero.sh, which runs on its operator's own terms.