loading…
Search for a command to run...
loading…
Discover and book theatre, shows, events, tours and experiences across 700+ cities worldwide on tickadoo® with real-time pricing and booking links.
Discover and book theatre, shows, events, tours and experiences across 700+ cities worldwide on tickadoo® with real-time pricing and booking links.
tickadoo® MCP Server brings live experience discovery to AI assistants through the Model Context Protocol (MCP). It gives compatible clients access to bookable theatre, shows, tours, attractions, and events across 680+ cities worldwide.
🌐 Languages: Landing page available in: EN · JA · 中文 · 한국어 · DE · FR · ES · IT · PT · tlhIngan Hol 🖖
tickadoo® MCPサーバーは、AIアシスタント(Claude、ChatGPT、Gemini等)から680以上の都市で13,000以上の体験(劇場、ツアー、アトラクション)を検索できます。APIキー不要、認証不要、即座に接続可能です。
tickadoo® MCP服务器让AI助手(Claude、ChatGPT、Gemini等)可以搜索680+城市中13,000+体验活动(剧院、旅游、景点)。无需API密钥,无需认证,即时连接。
tickadoo® MCP 서버는 AI 어시스턴트(Claude, ChatGPT, Gemini 등)에서 680+ 도시의 13,000+ 체험(극장, 투어, 관광명소)을 검색할 수 있습니다.
Current release: v1.4.2
_available_filters (10 fields: tag_counts, price_range, duration_range, audience, setting, physical, languages, tags, wheelchair, free_cancellation)_conversation_starters (contextual prompts based on result data)_related_searches (tag-based follow-up suggestions)_booking_urgency (conversion signals: TODAY availability, free cancellation, rating, wheelchair)_cross_sell, _intent_token, _accessibility| Tool | Description |
|---|---|
search_experiences |
Search 13,090+ experiences across 681 cities with 11 filters (audience, setting, wheelchair, physical level, duration, language, rating, cancellation, price, tags, category) + 6 sort options (incl. best_value) |
search_by_mood |
Search by emotional intent (romantic, relaxing, budget_friendly, rainy_day, adventurous, foodie) — maps mood to tags, audience, rating, setting, and price filters |
find_nearby_experiences |
Find experiences near lat/lng with the same 11 filters, configurable radius, and date filtering |
list_cities |
Browse supported cities with optional filtering and result limits |
check_availability |
Fast date-specific availability check for a single experience — returns matching slots, price-per-person, cheapest total for party_size, booking URL, and Ghost Checkout payload metadata |
get_experience_details |
Detailed availability, pricing, venue, and image information for a specific tickadoo experience by slug or booking path |
compare_experiences |
Side-by-side comparison of 2-5 experiences with winner callouts for best value, highest rated, most popular, and family fit |
get_whats_on_this_week |
7-day city planner with morning/afternoon/evening breakdown |
whats_on_tonight |
Tonight's experiences with start-time ranking and urgency signals |
get_last_minute |
Experiences starting within hours, sorted by soonest |
get_city_guide |
Curated city overview: top highlights, category mix, pricing ranges, best-for suggestions, insider tips |
get_travel_tips |
Local insider advice for 20 launch cities: transport, money, safety, culture, food, emergency numbers, quick phrases |
get_transfer_info |
Taxi, tube/metro, bus, and train transfer estimates from a city's default airport, station, or port to hotel coordinates |
get_family_day |
Family day planner with age-aware filtering and geographic clustering |
get_related_experiences |
Blended semantic + heuristic "you might also like" results for a given experience |
All tools expose MCP tool annotations for readOnlyHint, destructiveHint, and openWorldHint.
Every search response includes metadata keys designed for AI agents:
| Key | Description |
|---|---|
_available_filters |
12 fields: tags, audience, setting, price_range, duration_range, etc. |
_best_picks |
Auto-curated top 3: best_value, highest_rated, most_popular |
_price_tiers |
Budget/mid-range/premium grouping from result prices |
_group_summary |
Tag-based category breakdown (e.g. "8 Cruise, 6 GuidedTour, 5 Dining") |
_conversation_starters |
Context-aware prompts referencing actual products |
_related_searches |
Tag-based suggestions for narrowing results |
_next_step |
Recommended follow-up action |
get_experience_details adds: _booking_urgency, _cross_sell, _intent_token, _accessibility
| Resource | Description |
|---|---|
tickadoo://product-feed |
Machine-readable product feed in OpenAI Commerce Product Feed format |
Hosted MCP endpoint:
https://mcp.tickadoo.com/mcp
Example remote MCP config:
{
"mcpServers": {
"tickadoo": {
"url": "https://mcp.tickadoo.com/mcp"
}
}
}
Remote HTTP MCP (recommended — no API key, no local install):
claude mcp add --transport http tickadoo https://mcp.tickadoo.com/mcp
Or add as local stdio from npm:
claude mcp add tickadoo -- npx -y @tickadoo/mcp-server
See https://code.claude.com/docs/en/mcp for full Claude Code MCP docs.
Claude Dispatch uses the same remote MCP URL config:
{
"mcpServers": {
"tickadoo": {
"url": "https://mcp.tickadoo.com/mcp"
}
}
}
Add this to ~/.gemini/settings.json:
{
"mcpServers": {
"tickadoo": {
"httpUrl": "https://mcp.tickadoo.com/mcp"
}
}
}
Use Perplexity's custom remote connector:
tickadoohttps://mcp.tickadoo.com/mcpStreamable HTTPNoneAvailable to ChatGPT Pro, Business, Enterprise, and Edu plans:
tickadoohttps://mcp.tickadoo.com/mcpNoneThe same httpUrl config works in:
mcp.json).idx/mcp.json or .gemini/settings.json)Run from npm after publish:
npx -y @tickadoo/mcp-server
Example stdio config via npx:
{
"mcpServers": {
"tickadoo": {
"command": "npx",
"args": ["-y", "@tickadoo/mcp-server"]
}
}
}
Install and build:
npm install
npm run build
Run the stdio server directly:
node dist/index.js
Example stdio config:
{
"mcpServers": {
"tickadoo": {
"command": "node",
"args": ["/path/to/tickadoo-mcp/dist/index.js"]
}
}
}
| Command | Purpose |
|---|---|
npm run build |
Bundle the stdio server (dist/index.js) via esbuild |
npm test |
Run the unit test suite (vitest) |
npm run e2e:stdio |
Run the MCP smoke suite against the local stdio server |
npm run e2e:http |
Run the MCP smoke suite against an HTTP endpoint |
npm run dev:worker |
Run the production Worker locally via wrangler dev |
npm run deploy |
Deploy the main MCP Worker (wrangler deploy) |
npm run deploy:widgets |
Deploy the widgets Worker under widgets-worker/ |
Shared server logic lives in src/shared/* — API client, formatting, tool definitions, resource registration. Transport wrappers:
src/index.ts — local stdio transport (published on npm as @tickadoo/mcp-server)src/worker.ts — Cloudflare Worker (Hono + WebStandardStreamableHTTPServerTransport) serving https://mcp.tickadoo.comwidgets-worker/ — separate Cloudflare Worker serving embeddable widgets at https://widgets.tickadoo.comBoth workers deploy via Cloudflare Workers:
wrangler.jsonc → mcp.tickadoo.comwidgets-worker/wrangler.jsonc → widgets.tickadoo.comCI: .github/workflows/deploy-cf.yml runs on every push to main and deploys both workers.
| Variable | Default | Description |
|---|---|---|
TICKADOO_API_BASE |
https://api.tickadoo.com |
tickadoo API base URL |
TICKADOO_LOG_LEVEL |
info |
Logging verbosity: none, info, or debug |
TICKADOO_SITE_BASE |
https://www.tickadoo.com |
tickadoo website base URL |
tickadoo® is always lowercase. The tagline is What Do You Wanna Doo?® Please preserve lowercase in all integrations and documentation.
MIT — tickadoo Inc.
Добавь это в claude_desktop_config.json и перезапусти Claude Desktop.
{
"mcpServers": {
"tickadoo": {
"command": "npx",
"args": []
}
}
}