Command Palette

Search for a command to run...

UnylyUnyly
Весь каталог

Chrono

БесплатноНе проверен

Zero-dependency date & time MCP server — one file, pure Python stdlib, DST-aware. 7 tools: current time, convert, offset, natural-language parse, durations, mee

GitHubEmbed

Описание

Zero-dependency date & time MCP server — one file, pure Python stdlib, DST-aware. 7 tools: current time, convert, offset, natural-language parse, durations, meeting windows, format.

README

A zero-dependency date & time MCP server. One file, pure Python standard library — no mcp package, no pip install, no virtualenv. AI agents are notoriously bad at knowing today's date, doing timezone math, and remembering DST rules; this server gives them seven precise, tested tools for exactly that.

{"mcpServers": {"chrono": {"command": "python3", "args": ["/absolute/path/to/chrono_mcp.py"]}}}

Why

  • AI models don't know what time it is. Training data ends months before you ask; "today" is a guess. This server gives the real clock.
  • Timezone math is where agents hallucinate. "Is Shanghai 12 or 13 hours ahead of New York right now?" depends on when "now" is, because of DST. All conversions here are DST-aware via the IANA database.
  • Zero dependencies, by design. No supply chain, no version conflicts, nothing to audit. If you have Python 3.9+ you already have everything. The MCP protocol layer (JSON-RPC 2.0 over stdio) is implemented in the same file.

Tools

Tool What it does Example
get_current_time Real current date/time in any IANA zone (or city alias) {"timezone": "tokyo"}
convert_time Wall-clock conversion across zones with day-rollover detection 09:00 Shanghai → 21:00 New York (previous day)
time_offset Live UTC-offset difference between two zones (DST-aware) Shanghai → New York = -12h00m in September
parse_natural_time Free-form phrase → precise ISO timestamp "next tuesday 3pm", "3 days ago", "sep 10 3pm"
add_duration Signed duration arithmetic, incl. business days "+3 hours", "-2h30m", "3 business days"
meeting_windows Common meeting slots across timezones within local business hours Shanghai + London → 16:00 / 09:00
format_time strftime formatting with optional zone conversion %A, %B %d %Y %I:%M %p

City aliases work anywhere a timezone is accepted: tokyo, beijing, new york, san francisco, london, dubai, utc, and ~40 more.

Install

Any MCP client (Claude Desktop, Cursor, Windsurf, …)

Clone and point your client at the file:

{
  "mcpServers": {
    "chrono": {
      "command": "python3",
      "args": ["/absolute/path/to/chrono-mcp/chrono_mcp.py"]
    }
  }
}

Claude Desktop: claude mcp add chrono -- python3 /absolute/path/to/chrono-mcp/chrono_mcp.py

Requirements

  • Python 3.9+ with the zoneinfo module (standard on 3.9+; on Windows also install tzdata from PyPI if zone lookups fail).

Example phrases parse_natural_time understands

Phrase Result (Asia/Shanghai)
next tuesday 3pm the coming Tuesday, 15:00 local
tomorrow 09:00 tomorrow, 09:00 local
tonight today, 20:00 local
in 2 hours / 3 days ago relative to real current time
friday 9:00 the coming Friday
last friday the most recent Friday
2026-09-10 14:30 exact datetime
sep 10 3pm / september 10, 2026 2pm month-name dates

Example: meeting_windows

meeting_windows(timezones=["Asia/Shanghai", "Europe/London"], date="2026-09-02")
→ windows: [{ start_utc: 08:00, local_start: {Shanghai: 16:00, London: 09:00} }, …]

For a genuinely impossible pair (Shanghai + New York within 9–18 local), the server returns a clear No common window message with suggestions instead of guessing.

Run the tests

python3 -m unittest discover -s tests

47 tests, covering every tool plus end-to-end JSON-RPC over a real subprocess (initialize → tools/list → tools/call, error recovery, notification handling).

Protocol notes

Implements the stdio transport of the Model Context Protocol: newline-delimited JSON-RPC 2.0 with initialize, notifications/initialized, ping, tools/list, tools/call. Tool errors are returned as isError: true results (the server never crashes on bad input); unknown methods return -32601.

License

MIT

from github.com/w1977-0/chrono-mcp

Установка Chrono

У этого сервера нет опубликованного пакета — он собирается из исходников. Открой репозиторий и следуй инструкции в README.

▸ github.com/w1977-0/chrono-mcp

FAQ

Chrono MCP бесплатный?

Да, Chrono MCP бесплатный — установка в пару кликов через Unyly без оплаты.

Нужен ли API-ключ для Chrono?

Нет, Chrono работает без API-ключей и переменных окружения.

Chrono — hosted или self-hosted?

Self-hosted: сервер запускается локально на твоей машине командой из раздела установки.

Как установить Chrono в Claude Desktop, Claude Code или Cursor?

Открой Chrono на unyly.org, выбери вкладку своего клиента (Claude Desktop, Claude Code, Cursor) и нажми Install — конфиг сгенерируется автоматически, без правки JSON.

Похожие MCP

Fetch

Web content fetching and conversion for efficient LLM usage.

автор: Community

Roblox Studio

Enables AI coding tools to control Roblox Studio for workspace exploration, instance manipulation, and script management. It provides tools for playtesting, sce

paralovавтор: paralov

Opencode Omniroute Plugin

OpenCode plugin for the OmniRoute AI Gateway. Drives dynamic model discovery, /connect auth flow, and multi-instance OmniRoute providers via the official @openc

GitHub Actionsавтор: GitHub Actions

AWS KB Retrieval

Retrieval from AWS Knowledge Base using Bedrock Agent Runtime.

modelcontextprotocolавтор: modelcontextprotocol

Spring AI MCP Server

Provides auto-configuration for setting up an MCP server in Spring Boot applications.

автор: Community

llm-analysis-assistant

A very streamlined mcp client that supports calling and monitoring stdio/sse/streamableHttp, and can also view request responses through the /logs page. It also

xuzexin-hzавтор: xuzexin-hz

MCP-Agent

A simple, composable framework to build agents using Model Context Protocol by [LastMile AI](https://www.lastmileai.dev)

lastmile-aiавтор: lastmile-ai

Spring AI MCP Client

Provides auto-configuration for MCP client functionality in Spring Boot applications.

автор: Community

mcp.natoma.ai

A Hosted MCP Platform to discover, install, manage and deploy MCP servers by [Natoma Labs](https://www.natoma.ai)

автор: Community

MCPHub

Website to list high quality MCP servers and reviews by real users. Also provide online chatbot for popular LLM models with MCP server support.

автор: Community

Compare Chrono with

Не уверен что выбрать?

Найди свой стек за 60 секунд

Автор?

Embed-бейдж для README

Похожее

Все в категории ai