LLM Usage MCP Server

LLM Usage & Cost Tracker - your local-first spend watchdog English | 中文 Stop treating your LLM API bills like a scary horror movie you only look at through your fingers at the end of the month. Know what your LLM calls actually cost - across every provider, in one place, on your own machine. Ask your coding agent (MCP) or type a command (CLI).

People who work with deepseek, llm costs and local first and want it reachable from Claude, Cursor, VS Code, or another MCP client. The project is written in Python.

VERIFIED ACTIVE

LAST COMMIT 2026-09-17 · ★ 4 · #79 OF 120 MAINTAINED DEVOPS & MONITORING · VERIFIED 2026-09-18

MIT · Python servers · how we verify → /methodology

01 · Install LLM Usage

before you install - you'll need

Set ANTHROPIC_API_KEY, OPENAI_API_KEY, DEEPSEEK_API_KEY, DASHSCOPE_API_KEY, ANTHROPIC_BASE_URL before connecting.

Claude Code

claude mcp add zhaoyue722-llm-usage-mcp -- uvx llm-usage-mcp

Claude Desktop / Cursor / VS Code - add to config

{
  "mcpServers": {
    "zhaoyue722-llm-usage-mcp": {
      "command": "uvx",
      "args": [
        "llm-usage-mcp"
      ]
    }
  }
}

Same JSON for Cursor. For VS Code, rename the top-level key from `mcpServers` to `servers`.

Using another client? Same JSON, different key

Claude Desktop · mcpServers

Cursor · mcpServers

VS Code · servers

Windsurf · mcpServers

Zed · context_servers

Cline · mcpServers

Roo Code · mcpServers

Continue · mcpServers

LibreChat · mcpServers

Gemini CLI · mcpServers

Codex CLI · mcp_servers

Full setup guides: every client.

02 · Evidence

Security posture

What to check before giving this server access to your agent - from the registry, GitHub, and our own probes. We don't score safety; we show what's verifiable.

runs as local process (stdio) - runs on your machine with your user's permissions

license MIT - declared in the repository

pypi package llm-usage-mcp - check the name against the project README before installing (PyPI has no namespace ownership)

registry namespace io.github.zhaoyue722 is GitHub-verified and matches the repo owner

03 · What LLM Usage can do

Prose above is summarized from the project's README and registry record - no invented capabilities.

Latest releases

v0.1.2 · 2026-07-07

Official MCP Registry publish: adds server.json manifest and the mcp-name ownership marker. See CHANGELOG.md for details.

v0.1.1 · 2026-06-24

Fixed · Packaged installs now boot. v0.1.0 shipped without its Alembic migrations bundled in the wheel, so every pip install / uv tool install boot of llm-usage-mcp / llm-usage-proxy failed with alembic.ini not found.…

v0.1.0 · 2026-06-24

First public release of llm-usage-mcp - a local-first, multi-provider tool that captures LLM API spend and exposes it to coding agents via MCP and a CLI. · Highlights · Capture proxy (loopback-only 127.0.0.1:5525) for…

04 · Who maintains LLM Usage

LLM Usage is maintained by zhaoyue722. It's the only MCP server we track from this author; the repo dates to May 2026.

05 · Facts

repository: github.com/zhaoyue722/llm-usage-mcp
website: https://github.com/zhaoyue722/llm-usage-mcp
category: DevOps & monitoring - ranked #79 of 120 actively-maintained DevOps & monitoring servers as of 2026-09-18.
release cadence: 3 releases in the last 90 days (latest 2026-07-07)
registry: io.github.zhaoyue722/llm-usage-mcp (active, first published 2026-07-07 · 2 versions)
packages: pypi:llm-usage-mcp

06 · LLM Usage FAQ

What is LLM Usage?

Is LLM Usage still maintained?

Yes - as of 2026-09-18, its last commit was 2026-09-17 and it shipped 3 releases in the last 90 days. We re-verify nightly.

How do I install LLM Usage?

Run `uvx llm-usage-mcp`. The README documents 5 environment variables (ANTHROPIC_API_KEY, OPENAI_API_KEY, DEEPSEEK_API_KEY…) to set first. Set ANTHROPIC_API_KEY, OPENAI_API_KEY, DEEPSEEK_API_KEY, DASHSCOPE_API_KEY, ANTHROPIC_BASE_URL before connecting. You can also paste the ready-made client config above.

Does LLM Usage run locally?

Yes - it's a stdio server: it runs on your machine (via uvx) with your user's permissions. Your data stays local unless the server itself calls external APIs.

07 · Alternatives to LLM Usage

Alternatives to LLM Usage

Maintained DevOps & monitoring servers if LLM Usage isn't the fit.

Pairs well with

Servers that cover what LLM Usage doesn't - only shown when the pairing reason fits the companion.

More DevOps & monitoring MCP servers · VMware AIops · VMware Aria Operations · VMware Monitor · ReNoUn - Crypto Regime Classification · Agent Lens

More Python MCP servers · Nanobanana MCP Server · Pyobfus MCP · YouTube Summarize · Quant Context · Code Factory · see all