haomingkoo/japan-seasons-mcp

**Give your AI assistant live Japan travel data — cherry blossom forecasts, autumn leaves, fruit picking, flowers, festivals & more.**
*1,700+ spots. 17 tools. Live data from Japan Meteorological Corporation.*
**Live map** · **Hosted MCP endpoint** · **npm**
---
Screenshots
| Cherry blossom map | Spot detail & weather |
|---|---|
| | |
| Fruit picking | What's On in Japan |
|---|---|
| | |
| Flowers in season | Mobile |
|---|---|
| | |
---
The problem
You ask Claude, ChatGPT, or Gemini: *"When should I visit Kyoto for cherry blossoms?"*
You get "late March to early April" — the same answer every year, based on training data that has no idea what this winter was like.
The actual bloom date shifts by 2–3 weeks depending on temperatures. Tokyo peaked March 29 in 2024, March 22 in 2023, March 28 in 2022 — a 7-day spread in just two years. Your AI is guessing. This MCP server stops the guessing.
The fix
It connects Claude (or any MCP client) to live forecast data from the Japan Meteorological Corporation — the same data that powers SAKURA NAVI and Japan's major cherry blossom forecast sites. Bloom meters update daily at 9 AM JST; spot observations update when JMC reporters file.
Real bloom percentages. Real park names. Real weather.
---
Use it with AI assistants
There are two different ways an AI assistant can use Japan in Seasons:
1. **AI search / web browsing** can find and cite the public pages, text summaries, and JSON APIs. This is the lowest-friction path for people asking "latest sakura dates" in ChatGPT, Perplexity, Google AI Mode, etc.
2. **MCP tool use** only works after the user or workspace admin connects the MCP server inside an MCP-capable client. Finding this repo in search does not give the assistant permission to run MCP tools.
For AI search, use these crawlable sources:
- Latest sakura text summary: https://seasons.kooexperience.com/sakura-forecast.txt
- Sakura forecast JSON: https://seasons.kooexperience.com/api/sakura/forecast
- Interactive map and forecast pages: https://seasons.kooexperience.com
Remote MCP endpoint
No package install is needed when the client supports remote/streamable HTTP MCP. Add this as the MCP server/app/connector URL:
For ChatGPT Apps/Connectors, use:
- **Name:** Japan in Seasons
- **Description:** Use this for live Japan seasonal travel data: cherry blossom and sakura dates, autumn leaves, flowers, festivals, fruit picking, and weather. Best for current or date-specific Japan travel questions.
- **Connector URL:** https://seasons.kooexperience.com/mcp
Claude Desktop / Claude Code / stdio MCP clients
Add this to the client's MCP config:
Optional connection preferences supported by the hosted endpoint:
- dateStyle — friendly or iso
- temperatureUnit — celsius or fahrenheit
- includeCoordinates — true or false
- mapLanguage — english or japanese
Self-host
Optional durable cache for the hosted web map:
Point JAPAN_SEASONS_CACHE_DIR at a persistent volume, for example a Railway Volume mounted at /data. This stores the warmed national sakura/koyo all-spots JSON across restarts; without it, the app falls back to in-memory cache and still works.
Why ChatGPT may find it but refuse to use it
ChatGPT Search can discover japan-seasons-mcp as a web result, but web discovery is not the same as connecting an MCP server. If the chat only has web search enabled, it should cite the crawlable forecast pages or JSON API. To run tools like sakura_now, sakura_forecast, or sakura_spots, the MCP endpoint must first be added as a ChatGPT app/connector or configured in another MCP client.
---
What's covered
| Season | Data | Spots | Source |
|--------|------|-------|--------|
| Jan–Feb | Kawazu cherry (early deep-pink variety, Izu Peninsula) | 9 spots | JMC live |
| Jan–Mar | Plum blossoms | 8 spots | curated |
| Mar–May | Cherry blossom (sakura) | **1,012 parks & temples** | JMC live, daily |