Featured

Deploy OpenClaw in 60 seconds β€” 20% off logoDeploy OpenClaw in 60 seconds β€” 20% off

Launch OpenClaw on Hostinger in about 60 seconds and keep your agent live 24/7. Our referral link gives you 20% off, no coupon code needed.

Launch on Hostinger β†’
Run your Hermes agent on Hostinger, fully managed logoRun your Hermes agent on Hostinger, fully managed

Launch Hermes on Hostinger in one click, fully managed, no VPS knowledge needed. Use code ZACAARON10 for 10% off.

Launch on Hostinger β†’
Crawl and scrape any site into clean data, 10% off logoCrawl and scrape any site into clean data, 10% off

Firecrawl crawls and scrapes any site into clean markdown for your agent. Get 1,000 free credits, and new users get 10% off their first purchase.

Try Firecrawl free β†’
6,000+ web scrapers for your AI agent, start free logo6,000+ web scrapers for your AI agent, start free

Apify gives your agent live web data: 6,000+ prebuilt scrapers and actors, MCP-ready. Sign up free with $5 in usage credits.

Try Apify free β†’
One API to scrape, enrich, and extract the internet. logoOne API to scrape, enrich, and extract the internet.

Context.dev gives your agents a single API to scrape, enrich, and extract live web data β€” no proxies, no parsers, no maintenance.

Start building free β†’
SetupClaw: done-for-you OpenClaw for founders & exec teams logoSetupClaw: done-for-you OpenClaw for founders & exec teams

White-glove OpenClaw for founders and exec teams (4–50+ employees): we install, harden, integrate your tools, and maintain it β€” secured from day one.

Get it set up for you β†’
SEO data APIs for your agent, $1 free credit logoSEO data APIs for your agent, $1 free credit

DataForSEO gives your agent live access to SERP results, keyword data, backlinks, and on-page SEO data through one API. New accounts get a $1 credit, good for up to 20,000 keyword or backlink lookups.

Try DataForSEO free β†’
Reach 48,000+ AI builders

A flat monthly placement in front of developers actively installing AI tools. No lock-in, cancel anytime.

Advertise here β†’

Works with

Claude CodeClaude DesktopCursorVS CodeClineCodex CLIOpenClaw+ any MCP client

Install to Claude Code

This server doesn't publish a one-line install command. Follow the setup in the source repository.

Summary

nightlife](https://glama.ai/mcp/servers/alcylu/nightlife-mcp/badges/score.svg)](https://glama.ai/mcp/servers/alcylu/nightlife-mcp) πŸ“‡ ☁️ - MCP server for Tokyo nightlife event discovery, venue search, performer info, AI recommendations, and VIP table booking.

README.md

nightlife-mcp

<a href="https://glama.ai/mcp/servers/alcylu/nightlife-mcp"> <img alt="nightlife-mcp MCP server" src="https://glama.ai/mcp/servers/alcylu/nightlife-mcp/badge" /> </a>

MCP server for nightlife event discovery backed by Supabase.

Quick Start

Production endpoint: https://api.nightlife.dev/mcp

  1. Get a free API key at nightlife.dev
  2. Add to Claude Desktop config (claude_desktop_config.json):
{
  "mcpServers": {
    "nightlife": {
      "url": "https://api.nightlife.dev/mcp",
      "headers": {
        "x-api-key": "YOUR_API_KEY"
      }
    }
  }
}

For curl, TypeScript SDK, and other clients, see CLIENT_SETUP.md.

Implemented (v0.3)

  • search_events
  • get_tonight
  • get_event_details
  • search_venues
  • get_venue_info
  • search_performers
  • get_performer_info
  • log_unmet_request
  • create_vip_booking_request
  • get_vip_booking_status
  • get_vip_table_availability
  • get_vip_table_chart
  • get_recommendations (v0.2, behind MCP_ENABLE_RECOMMENDATIONS=true)
  • Streamable HTTP endpoint with API-key middleware
  • Structured tool output schemas (outputSchema)
  • Deterministic tool error payloads with stable error codes
  • Runtime request/tool metrics exposed at /health

Prerequisites

  • Node.js 18+
  • Supabase project URL + service role key

Setup

cp .env.example .env
npm install

Set env vars in .env:

  • SUPABASE_URL
  • SUPABASE_SERVICE_ROLE_KEY

Optional:

  • DEFAULT_CITY (default: tokyo)
  • MCP_TOP_LEVEL_CITIES (default example: tokyo,san-francisco; controls available_cities in unsupported-city responses)
  • DEFAULT_COUNTRY_CODE (default: JP)
  • NIGHTLIFE_BASE_URL (default: https://nightlifetokyo.com)
  • MCP_HTTP_REQUIRE_API_KEY (default: true)
  • MCP_HTTP_USE_DB_KEYS (default: true)
  • MCP_HTTP_ALLOW_ENV_KEY_FALLBACK (default: true)
  • MCP_HTTP_API_KEYS (comma-separated legacy fallback keys)
  • MCP_ENABLE_RECOMMENDATIONS (default: false; enables get_recommendations)

DB API Key Mode (Recommended)

HTTP authentication can use persistent API keys from Supabase plus quota tracking.

1) Run SQL migration in your Supabase SQL editor:

-- copy file contents from:
-- supabase/migrations/20260219094000_mcp_api_keys.sql

2) Ensure DB auth flags are enabled in .env:

MCP_HTTP_REQUIRE_API_KEY=true
MCP_HTTP_USE_DB_KEYS=true
MCP_HTTP_ALLOW_ENV_KEY_FALLBACK=true

3) Create an API key record:

npm run key:create -- --name claude-desktop --tier starter --daily-quota 1000 --minute-quota 60

This prints the raw api_key once. Save it securely and use it in MCP HTTP calls.

If the DB RPC is unavailable, fallback to MCP_HTTP_API_KEYS works only when MCP_HTTP_ALLOW_ENV_KEY_FALLBACK=true.

Concierge Unmet Request Backlog

Public concierge flows can log unsupported user intents to Supabase.

Run migration:

-- copy file contents from:
-- supabase/migrations/20260226_concierge_unmet_requests.sql

Then call MCP tool log_unmet_request when no good answer exists from available nightlife data.

VIP Booking Phase 1

VIP table booking submission and status tracking are backed by Supabase.

Run migration:

-- copy file contents from:
-- supabase/migrations/20260227143000_vip_phase1_requests_and_queue.sql
-- supabase/migrations/20260228124500_add_vip_booking_enabled_to_venues.sql
-- and if already deployed before 2026-02-28:
-- supabase/migrations/20260228111000_vip_outward_language_defaults.sql
-- supabase/migrations/20260301010000_vip_table_availability_chart.sql
-- supabase/migrations/20260301114000_vip_table_chart_storage_bucket.sql
-- supabase/migrations/20260303093000_vip_dashboard_admin_edits.sql

Then call MCP tools:

  • create_vip_booking_request
  • get_vip_booking_status
  • get_vip_table_availability (read per-day table availability by venue/date range)
  • get_vip_table_chart (read structured table chart with optional per-date status overlay and optional layout_image_url)

Conversation policy for create_vip_booking_request:

  • Confirm booking date/time in venue local time before submitting.
  • Use dual-date confirmation wording, especially for late-night arrivals (00:00-05:59).
  • Required confirmation template:
  • Just to confirm: you want a table for [Night Day] night ([Night Date]), arriving around [Time] on [Arrival Day], [Arrival Date] ([Timezone]). I'll submit that as [Night Day] night with [Time] arrival. Is that correct?
  • If the user gives a time like 2am without a day, ask:
  • Do you mean 2:00 AM after Thursday night (Friday morning), or after Friday night (Saturday morning)?
  • If the user changes day, regenerate confirmation before submission.

Ops-tier sessions also have internal queue tools:

  • list_vip_reservations (all outstanding reservations; default statuses: submitted, in_review, confirmed)
  • list_vip_requests_for_alerting (due alerts only)
  • mark_vip_request_alert_sent
  • claim_vip_request_after_ack
  • update_vip_booking_status (set confirmed/rejected/cancelled with audit event)
  • upsert_vip_venue_tables (write venue table definitions + chart coordinates + optional table notes + optional layout_image_url)
  • upsert_vip_table_availability (write per-date table statuses)
  • upload_vip_table_chart_image (upload chart image to storage and attach layout_image_url to venue table metadata)

To discover bookable venues first, use:

  • search_venues with vip_booking_supported_only=true
  • or get_venue_info and check vip_booking_supported

For internal venue-booking workers, claim queue tasks via DB function:

  • public.claim_next_vip_agent_task(p_agent_id text)

VIP Ops Dashboard

Internal dashboard routes:

  • GET /ops/login
  • POST /ops/login
  • POST /ops/logout
  • GET /ops/vip-dashboard

Internal admin API routes (cookie-session protected):

  • GET /api/v1/admin/vip-bookings
  • GET /api/v1/admin/vip-bookings/:id
  • PATCH /api/v1/admin/vip-bookings/:id

Required env vars:

  • VIP_DASHBOARD_ADMINS (comma-separated username:password pairs)

Optional env vars:

  • VIP_DASHBOARD_SESSION_TTL_MINUTES (default: 720)
  • VIP_DASHBOARD_SESSION_COOKIE_NAME (default: vip_dashboard_session)

Run

Stdio (local desktop clients):

npm run dev

Streamable HTTP:

npm run dev:http

For production build:

npm run build
npm start

For HTTP in production:

npm run start:http

Authenticated production smoke check:

NLT_API_KEY=... npm run smoke:prod:auth

Optional env overrides:

  • NLT_MCP_URL (default: https://api.nightlife.dev/mcp)
  • NLT_REST_BASE_URL (default: https://api.nightlife.dev/api/v1)
  • NLT_SMOKE_CITY (default: tokyo)

Debug web UI for recommendations:

MCP_ENABLE_RECOMMENDATIONS=true npm run dev:http
# open http://127.0.0.1:3000/debug/recommendations

Notes

  • Date handling supports tonight, this_weekend, YYYY-MM-DD, and YYYY-MM-DD/YYYY-MM-DD.
  • get_recommendations returns up to 10 diverse modal slots with dynamic city-aware fallback.
  • Venue and performer tools include upcoming events snapshots.
  • log_unmet_request writes unresolved user asks to public.concierge_unmet_requests.
  • VIP phase 1 writes booking submissions to public.vip_booking_requests and worker queue tasks to public.vip_agent_tasks.
  • VIP inventory writes table definitions to public.vip_venue_tables and date-specific statuses to public.vip_table_availability.
  • search_venues and get_venue_info include vip_booking_supported so clients can show exactly which venues accept VIP booking submissions.
  • vip_booking_supported is sourced from public.venues.vip_booking_enabled (separate from guest_list_enabled).
  • create_vip_booking_request only accepts venues where vip_booking_supported=true.
  • City handling is backed by public.cities (slug, timezone, and service-day cutoff).
  • Supported top-level cities are environment-configurable (for example tokyo and san-francisco) while Tokyo can remain the default.
  • Stdio transport: no API key check.
  • HTTP transport (/mcp): API key required by default (MCP_HTTP_REQUIRE_API_KEY=true).
  • API key headers:
  • Authorization: Bearer <key>
  • x-api-key: <key>
  • HTTP responses include key tier/source and rate-limit headers when DB-backed auth is active:
  • X-API-Key-Tier
  • X-API-Key-Source
  • X-RateLimit-Daily-Limit
  • X-RateLimit-Daily-Remaining
  • X-RateLimit-Minute-Limit
  • X-RateLimit-Minute-Remaining
  • Health endpoint: /health.
  • Debug page for manual tool testing: /debug/recommendations.
  • Tool errors are returned as JSON text payloads in result.content[0].text:
  • INVALID_DATE_FILTER
  • INVALID_EVENT_ID
  • UNSUPPORTED_EVENT_ID
  • EVENT_NOT_FOUND
  • INVALID_VENUE_ID
  • VENUE_NOT_FOUND
  • INVALID_PERFORMER_ID
  • PERFORMER_NOT_FOUND
  • INVALID_BOOKING_REQUEST
  • BOOKING_REQUEST_NOT_FOUND
  • BOOKING_STATUS_UPDATE_FAILED
  • VIP_TASK_NOT_AVAILABLE
  • VIP_ALERT_UPDATE_FAILED
  • VIP_CLAIM_FAILED
  • INVALID_REQUEST
  • REQUEST_WRITE_FAILED
  • DB_QUERY_FAILED
  • INTERNAL_ERROR

Contributing & CI

See CONTRIBUTING.md for the workflow and rules. AI agents read AGENTS.md. CLAUDE.md is the architecture source of truth.

See related servers & alternatives β†’

Related MCP servers

Browse all β†’

Related guides

Hand-picked reading to help you choose and use Search servers.