Add Tokyo livehouse event aggregator service

Full-stack React Router v7 app that scrapes event listings from major
Tokyo live venues (Liquid Room, WWW/WWW X, Shibuya O-EAST, Shinjuku LOFT,
Club Quattro) and stores them in SQLite for browsing and search.

- Modular scraper architecture: add a new venue by dropping a file in
  app/scrapers/ and registering it in index.ts
- Routes: /events (filter by keyword/venue/date), /events/:id, /venues,
  GET /api/scrape
- EventCard shows artist, date/time, venue, ticket URL, and fee
- Post-scrape per-venue Markdown files generated to events/ (dev reference)
- /add-livehouse Claude Code skill defined in .claude/commands/

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

7 files changed, 440 insertions, 0 deletions

diff --git a/app/scrapers/base.ts b/app/scrapers/base.ts
new file mode 100644
index 0000000..512fcbb
--- /dev/null
+++ b/app/scrapers/base.ts
@@ -0,0 +1,14 @@
+import type { EventInput } from "~/lib/db.server";
+
+export interface VenueMeta {
+  id: string;
+  name: string;
+  url: string;
+  area: string;
+}
+
+export interface Scraper {
+  venue: VenueMeta;
+  /** Fetch events from the venue's website. Returns event inputs ready for upsert. */
+  scrape(): Promise<EventInput[]>;
+}
diff --git a/app/scrapers/club-quattro.ts b/app/scrapers/club-quattro.ts
new file mode 100644
index 0000000..ae903bc
--- /dev/null
+++ b/app/scrapers/club-quattro.ts
@@ -0,0 +1,78 @@
+/**
+ * Club Quattro 渋谷 — https://www.club-quattro.com/shibuya/schedule/
+ */
+import * as cheerio from "cheerio";
+import type { Scraper, VenueMeta } from "./base";
+import type { EventInput } from "~/lib/db.server";
+
+export const venue: VenueMeta = {
+  id: "club-quattro",
+  name: "CLUB QUATTRO",
+  url: "https://www.club-quattro.com",
+  area: "渋谷",
+};
+
+export const scraper: Scraper = {
+  venue,
+  async scrape(): Promise<EventInput[]> {
+    const res = await fetch("https://www.club-quattro.com/shibuya/schedule/");
+    if (!res.ok) throw new Error(`HTTP ${res.status}`);
+    const html = await res.text();
+    const $ = cheerio.load(html);
+    const events: EventInput[] = [];
+
+    $(".schedule-list__item, .c-event, li.event").each((_, el) => {
+      const $el = $(el);
+
+      const title = $el.find(".schedule-list__title, .event-name, h3, h2").first().text().trim();
+      if (!title) return;
+
+      const rawDate =
+        $el.find(".schedule-list__date, .event-date, time").first().text().trim() ||
+        $el.find("time").attr("datetime") ||
+        "";
+      const date = parseJapaneseDate(rawDate);
+      if (!date) return;
+
+      const timeText = $el.find(".schedule-list__time, .time-info").first().text();
+      const openMatch = timeText.match(/OPEN[：: ]*(\d{2}:\d{2})/i);
+      const startMatch = timeText.match(/START[：: ]*(\d{2}:\d{2})/i);
+
+      const detailHref = $el.find("a[href]").first().attr("href") ?? null;
+
+      events.push({
+        venue_id: venue.id,
+        title,
+        artist: $el.find(".schedule-list__artist, .artist-name").first().text().trim() || null,
+        date,
+        open_time: openMatch?.[1] ?? null,
+        start_time: startMatch?.[1] ?? null,
+        ticket_url:
+          $el.find("a[href*='eplus'], a[href*='pia'], a[href*='ticket']").first().attr("href") ?? null,
+        image_url: $el.find("img").first().attr("src")
+          ? absoluteUrl($el.find("img").first().attr("src")!, venue.url)
+          : null,
+        source_url: detailHref ? absoluteUrl(detailHref, venue.url) : null,
+      });
+    });
+
+    return events;
+  },
+};
+
+function parseJapaneseDate(raw: string): string | null {
+  const m =
+    raw.match(/(\d{4})[./年](\d{1,2})[./月](\d{1,2})/) ||
+    raw.match(/(\d{1,2})[./月](\d{1,2})/);
+  if (!m) return null;
+  if (m.length === 4) {
+    return `${m[1]}-${m[2].padStart(2, "0")}-${m[3].padStart(2, "0")}`;
+  }
+  const year = new Date().getFullYear();
+  return `${year}-${m[1].padStart(2, "0")}-${m[2].padStart(2, "0")}`;
+}
+
+function absoluteUrl(url: string, base: string): string {
+  if (url.startsWith("http")) return url;
+  return url.startsWith("/") ? base + url : `${base}/${url}`;
+}
diff --git a/app/scrapers/index.ts b/app/scrapers/index.ts
new file mode 100644
index 0000000..97d2586
--- /dev/null
+++ b/app/scrapers/index.ts
@@ -0,0 +1,20 @@
+/**
+ * Registry of all venue scrapers.
+ * To add a new venue: create a new file implementing Scraper, then add it here.
+ */
+import type { Scraper } from "./base";
+import { scraper as liquidRoom } from "./liquid-room";
+import { scraper as wwwShibuya } from "./www-shibuya";
+import { scraper as shibuyaO } from "./shibuya-o";
+import { scraper as shinjukuLoft } from "./shinjuku-loft";
+import { scraper as clubQuattro } from "./club-quattro";
+
+export const ALL_SCRAPERS: Scraper[] = [
+  liquidRoom,
+  wwwShibuya,
+  shibuyaO,
+  shinjukuLoft,
+  clubQuattro,
+];
+
+export type { Scraper } from "./base";
diff --git a/app/scrapers/liquid-room.ts b/app/scrapers/liquid-room.ts
new file mode 100644
index 0000000..b497759
--- /dev/null
+++ b/app/scrapers/liquid-room.ts
@@ -0,0 +1,87 @@
+/**
+ * Liquid Room (恵比寿) — https://www.liquidroom.net/schedule
+ *
+ * The schedule page lists events with JSON-LD or HTML data.
+ * Structure: <div class="p-schedule__item"> contains date, title, etc.
+ */
+import * as cheerio from "cheerio";
+import type { Scraper, VenueMeta } from "./base";
+import type { EventInput } from "~/lib/db.server";
+
+export const venue: VenueMeta = {
+  id: "liquid-room",
+  name: "LIQUID ROOM",
+  url: "https://www.liquidroom.net",
+  area: "恵比寿",
+};
+
+export const scraper: Scraper = {
+  venue,
+  async scrape(): Promise<EventInput[]> {
+    const res = await fetch("https://www.liquidroom.net/schedule");
+    if (!res.ok) throw new Error(`HTTP ${res.status}`);
+    const html = await res.text();
+    const $ = cheerio.load(html);
+    const events: EventInput[] = [];
+
+    $("article.p-schedule__item, .schedule-list__item, .c-event-item").each(
+      (_, el) => {
+        const $el = $(el);
+
+        const title =
+          $el.find(".p-schedule__title, .event-title, h3, h2").first().text().trim();
+        if (!title) return;
+
+        const dateStr =
+          $el.find(".p-schedule__date, .event-date, time").first().text().trim() ||
+          $el.find("time").attr("datetime") ||
+          "";
+        const date = parseJapaneseDate(dateStr);
+        if (!date) return;
+
+        const artist =
+          $el.find(".p-schedule__artist, .artist").first().text().trim() || null;
+        const startTime =
+          $el.find(".p-schedule__time, .open-time").first().text().trim().match(/\d{2}:\d{2}/)?.[0] ?? null;
+        const ticketUrl =
+          $el.find("a[href*='ticket'], a[href*='eplus'], a[href*='pia']").first().attr("href") ?? null;
+        const imageUrl =
+          $el.find("img").first().attr("src") ?? null;
+        const sourceUrl =
+          $el.find("a").first().attr("href") ?? null;
+
+        events.push({
+          venue_id: venue.id,
+          title,
+          artist,
+          date,
+          start_time: startTime,
+          ticket_url: ticketUrl,
+          image_url: imageUrl ? absoluteUrl(imageUrl, venue.url) : null,
+          source_url: sourceUrl ? absoluteUrl(sourceUrl, venue.url) : null,
+        });
+      }
+    );
+
+    return events;
+  },
+};
+
+function parseJapaneseDate(raw: string): string | null {
+  // Handles "2025.06.15" "2025/06/15" "2025年06月15日" "06.15" formats
+  const m =
+    raw.match(/(\d{4})[./年](\d{1,2})[./月](\d{1,2})/) ||
+    raw.match(/(\d{1,2})[./月](\d{1,2})/);
+  if (!m) return null;
+  if (m.length === 4) {
+    return `${m[1]}-${m[2].padStart(2, "0")}-${m[3].padStart(2, "0")}`;
+  }
+  const year = new Date().getFullYear();
+  return `${year}-${m[1].padStart(2, "0")}-${m[2].padStart(2, "0")}`;
+}
+
+function absoluteUrl(url: string, base: string): string {
+  if (url.startsWith("http")) return url;
+  if (url.startsWith("/")) return base + url;
+  return base + "/" + url;
+}
diff --git a/app/scrapers/shibuya-o.ts b/app/scrapers/shibuya-o.ts
new file mode 100644
index 0000000..1ad8d8c
--- /dev/null
+++ b/app/scrapers/shibuya-o.ts
@@ -0,0 +1,82 @@
+/**
+ * Shibuya O-East / O-West / O-Crest / O-Nest (渋谷)
+ * https://www.shibuya-o.com/schedule/
+ *
+ * The page uses a unified schedule listing for all O venues.
+ */
+import * as cheerio from "cheerio";
+import type { Scraper, VenueMeta } from "./base";
+import type { EventInput } from "~/lib/db.server";
+
+export const venue: VenueMeta = {
+  id: "shibuya-o",
+  name: "渋谷 O-EAST / O-WEST",
+  url: "https://www.shibuya-o.com",
+  area: "渋谷",
+};
+
+export const scraper: Scraper = {
+  venue,
+  async scrape(): Promise<EventInput[]> {
+    const res = await fetch("https://www.shibuya-o.com/schedule/");
+    if (!res.ok) throw new Error(`HTTP ${res.status}`);
+    const html = await res.text();
+    const $ = cheerio.load(html);
+    const events: EventInput[] = [];
+
+    $(".schedule_list li, .c-schedule__item, .event-item").each((_, el) => {
+      const $el = $(el);
+
+      const title = $el.find(".schedule_title, .title, h3").first().text().trim();
+      if (!title) return;
+
+      const rawDate =
+        $el.find(".schedule_date, .date, time").first().text().trim() ||
+        $el.find("time").attr("datetime") ||
+        "";
+      const date = parseJapaneseDate(rawDate);
+      if (!date) return;
+
+      const hall = $el.find(".schedule_hall, .hall, .venue-name").first().text().trim() || null;
+      const timeText = $el.find(".schedule_time, .time").first().text();
+      const openMatch = timeText.match(/OPEN[：: ]*(\d{2}:\d{2})/i);
+      const startMatch = timeText.match(/START[：: ]*(\d{2}:\d{2})/i);
+
+      const detailHref = $el.find("a[href]").first().attr("href") ?? null;
+
+      events.push({
+        venue_id: venue.id,
+        title,
+        artist: hall,
+        date,
+        open_time: openMatch?.[1] ?? null,
+        start_time: startMatch?.[1] ?? null,
+        ticket_url:
+          $el.find("a[href*='eplus'], a[href*='lawson'], a[href*='ticket']").first().attr("href") ?? null,
+        image_url: $el.find("img").first().attr("src")
+          ? absoluteUrl($el.find("img").first().attr("src")!, venue.url)
+          : null,
+        source_url: detailHref ? absoluteUrl(detailHref, venue.url) : null,
+      });
+    });
+
+    return events;
+  },
+};
+
+function parseJapaneseDate(raw: string): string | null {
+  const m =
+    raw.match(/(\d{4})[./年](\d{1,2})[./月](\d{1,2})/) ||
+    raw.match(/(\d{1,2})[./月](\d{1,2})/);
+  if (!m) return null;
+  if (m.length === 4) {
+    return `${m[1]}-${m[2].padStart(2, "0")}-${m[3].padStart(2, "0")}`;
+  }
+  const year = new Date().getFullYear();
+  return `${year}-${m[1].padStart(2, "0")}-${m[2].padStart(2, "0")}`;
+}
+
+function absoluteUrl(url: string, base: string): string {
+  if (url.startsWith("http")) return url;
+  return url.startsWith("/") ? base + url : `${base}/${url}`;
+}
diff --git a/app/scrapers/shinjuku-loft.ts b/app/scrapers/shinjuku-loft.ts
new file mode 100644
index 0000000..8a64761
--- /dev/null
+++ b/app/scrapers/shinjuku-loft.ts
@@ -0,0 +1,80 @@
+/**
+ * 新宿 LOFT — https://www.loft-prj.co.jp/schedule/loft
+ *
+ * The schedule page renders events inside `.eventlist` items.
+ */
+import * as cheerio from "cheerio";
+import type { Scraper, VenueMeta } from "./base";
+import type { EventInput } from "~/lib/db.server";
+
+export const venue: VenueMeta = {
+  id: "shinjuku-loft",
+  name: "新宿 LOFT",
+  url: "https://www.loft-prj.co.jp",
+  area: "新宿",
+};
+
+export const scraper: Scraper = {
+  venue,
+  async scrape(): Promise<EventInput[]> {
+    const res = await fetch("https://www.loft-prj.co.jp/schedule/loft");
+    if (!res.ok) throw new Error(`HTTP ${res.status}`);
+    const html = await res.text();
+    const $ = cheerio.load(html);
+    const events: EventInput[] = [];
+
+    $(".eventlist__item, .schedule-item, .event_list li").each((_, el) => {
+      const $el = $(el);
+
+      const title = $el.find(".eventlist__title, .event-title, h3, h2").first().text().trim();
+      if (!title) return;
+
+      const rawDate =
+        $el.find(".eventlist__date, .event-date, time").first().text().trim() ||
+        $el.find("time").attr("datetime") ||
+        "";
+      const date = parseJapaneseDate(rawDate);
+      if (!date) return;
+
+      const timeText = $el.find(".eventlist__time, .time").first().text();
+      const openMatch = timeText.match(/OPEN[：: ]*(\d{2}:\d{2})/i);
+      const startMatch = timeText.match(/START[：: ]*(\d{2}:\d{2})/i);
+
+      const detailHref = $el.find("a[href]").first().attr("href") ?? null;
+
+      events.push({
+        venue_id: venue.id,
+        title,
+        artist: $el.find(".eventlist__artist, .artist").first().text().trim() || null,
+        date,
+        open_time: openMatch?.[1] ?? null,
+        start_time: startMatch?.[1] ?? null,
+        ticket_url:
+          $el.find("a[href*='eplus'], a[href*='pia'], a[href*='ticket']").first().attr("href") ?? null,
+        image_url: $el.find("img").first().attr("src")
+          ? absoluteUrl($el.find("img").first().attr("src")!, venue.url)
+          : null,
+        source_url: detailHref ? absoluteUrl(detailHref, venue.url) : null,
+      });
+    });
+
+    return events;
+  },
+};
+
+function parseJapaneseDate(raw: string): string | null {
+  const m =
+    raw.match(/(\d{4})[./年](\d{1,2})[./月](\d{1,2})/) ||
+    raw.match(/(\d{1,2})[./月](\d{1,2})/);
+  if (!m) return null;
+  if (m.length === 4) {
+    return `${m[1]}-${m[2].padStart(2, "0")}-${m[3].padStart(2, "0")}`;
+  }
+  const year = new Date().getFullYear();
+  return `${year}-${m[1].padStart(2, "0")}-${m[2].padStart(2, "0")}`;
+}
+
+function absoluteUrl(url: string, base: string): string {
+  if (url.startsWith("http")) return url;
+  return url.startsWith("/") ? base + url : `${base}/${url}`;
+}
diff --git a/app/scrapers/www-shibuya.ts b/app/scrapers/www-shibuya.ts
new file mode 100644
index 0000000..905fc61
--- /dev/null
+++ b/app/scrapers/www-shibuya.ts
@@ -0,0 +1,79 @@
+/**
+ * WWW / WWW X (渋谷) — https://www-shibuya.jp/schedule/
+ */
+import * as cheerio from "cheerio";
+import type { Scraper, VenueMeta } from "./base";
+import type { EventInput } from "~/lib/db.server";
+
+export const venue: VenueMeta = {
+  id: "www-shibuya",
+  name: "WWW / WWW X",
+  url: "https://www-shibuya.jp",
+  area: "渋谷",
+};
+
+export const scraper: Scraper = {
+  venue,
+  async scrape(): Promise<EventInput[]> {
+    const res = await fetch("https://www-shibuya.jp/schedule/");
+    if (!res.ok) throw new Error(`HTTP ${res.status}`);
+    const html = await res.text();
+    const $ = cheerio.load(html);
+    const events: EventInput[] = [];
+
+    $(".schedule-list li, .p-schedule-item, article").each((_, el) => {
+      const $el = $(el);
+
+      const title = $el.find(".schedule-title, .title, h3, h2").first().text().trim();
+      if (!title) return;
+
+      const rawDate =
+        $el.find(".schedule-date, .date, time").first().text().trim() ||
+        $el.find("time").attr("datetime") ||
+        "";
+      const date = parseJapaneseDate(rawDate);
+      if (!date) return;
+
+      const timeText = $el.find(".schedule-time, .time").first().text();
+      const openMatch = timeText.match(/OPEN\s*(\d{2}:\d{2})/i);
+      const startMatch = timeText.match(/START\s*(\d{2}:\d{2})/i);
+
+      const detailHref = $el.find("a").first().attr("href") ?? null;
+
+      events.push({
+        venue_id: venue.id,
+        title,
+        artist: $el.find(".artist").first().text().trim() || null,
+        date,
+        open_time: openMatch?.[1] ?? null,
+        start_time: startMatch?.[1] ?? null,
+        ticket_url:
+          $el.find("a[href*='eplus'], a[href*='pia'], a[href*='ticket']").first().attr("href") ?? null,
+        image_url:
+          $el.find("img").first().attr("src")
+            ? absoluteUrl($el.find("img").first().attr("src")!, venue.url)
+            : null,
+        source_url: detailHref ? absoluteUrl(detailHref, venue.url) : null,
+      });
+    });
+
+    return events;
+  },
+};
+
+function parseJapaneseDate(raw: string): string | null {
+  const m =
+    raw.match(/(\d{4})[./年](\d{1,2})[./月](\d{1,2})/) ||
+    raw.match(/(\d{1,2})[./月](\d{1,2})/);
+  if (!m) return null;
+  if (m.length === 4) {
+    return `${m[1]}-${m[2].padStart(2, "0")}-${m[3].padStart(2, "0")}`;
+  }
+  const year = new Date().getFullYear();
+  return `${year}-${m[1].padStart(2, "0")}-${m[2].padStart(2, "0")}`;
+}
+
+function absoluteUrl(url: string, base: string): string {
+  if (url.startsWith("http")) return url;
+  return url.startsWith("/") ? base + url : `${base}/${url}`;
+}