DEV Community: atu ino

Weekly Japanese Anime Figure & Collectibles Market Report (Week 40, 2026)

atu ino — Sun, 04 Oct 2026 21:32:12 +0000

This Week's Japanese Hobby Market — Key Takeaways

Fresh weekly price data is now available for Japanese anime figures, plastic models (Gunpla), trading cards, and collectibles. Week 40 covers 7 major Japanese marketplaces with 1,000+ price records from actual sold and listing data.

📊 What's in This Week's Dataset

Marketplace	Sample Items	Price Range (JPY)	Key Trends
Yahoo! Auctions Japan	400+	¥500 – ¥150,000	High-end figures trending upward
Mercari Japan	350+	¥300 – ¥80,000	Volume leader for mid-range items
Mandarake	200+	¥1,000 – ¥200,000	Rare/vintage premium holding strong
Suruga-ya	150+	¥800 – ¥120,000	Consistent retailer pricing
Rakuten Ichiba	100+	¥2,000 – ¥90,000	New release pre-orders active
Amazon Japan	80+	¥1,500 – ¥60,000	Prime-eligible fast movers
Yahoo! Shopping	60+	¥1,000 – ¥70,000	Bundle deals emerging

🎯 Top Movers This Week

Rising:

Vintage 1980s–90s figures: Steady collector demand
Trading card singles (holos/alt-arts): Volatile but upward bias
Limited edition Gunpla (MG/RG/PG): +6–10% on re-releases

Stable/Declining:

Common prize figures (¥1,000–3,000): Oversupply, price compression
Standard edition Nendoroids: Flat, ample stock

📈 Why This Data Matters

If you're a reseller, collector, importer, or researcher tracking Japanese secondhand markets, you need reliable price benchmarks — not guesswork from a single listing. This weekly CSV gives you:

Actual sold prices (not asking prices) from completed transactions
Condition-graded data (Mint/Opened/Box Damaged/etc.)
Brand, series, character, release date metadata for filtering
Historical trend columns for week-over-week comparison

🔗 Get the Data

Product	Price	Description	Link
Free Sample (30 rows, 7 marketplaces)	FREE	Frozen snapshot — try before you buy	https://atushi5.gumroad.com/l/kutuxe
Full Weekly Dataset (1,000+ rows, weekly updates)	$39.99	Complete CSV with all marketplaces, historical trends, condition grading	https://atushi5.gumroad.com/l/agyhq
Weekly Deep-Dive Market Report	$10	Analysis, charts, top movers, resale strategy notes	https://atushi5.gumroad.com/l/qdyyyi

🛠 Technical Details

Format: CSV (UTF-8), compatible with Excel, Python/pandas, Google Sheets, Power BI, Tableau
No API key required — direct download after purchase
Updated every Monday with the prior week's completed transactions
Historical archive included — build your own price models

📚 Use Cases

Resale arbitrage: Buy undervalued, sell at market rate
Collection valuation: Know what your shelf is worth
Market research: Track Japanese hobby demand cycles
AI/ML training: Clean, labeled price data for forecasting models
Import/export: Calibrate buy prices for Japanese sourcing

🤖 Related Apify Scrapers

The data behind this report comes from Apify actors I built for Japanese secondhand market research:

This summary is generated from the weekly Japanese Hobby & Collectibles Market Price Dataset. Data sources: Yahoo! Auctions Japan, Mercari Japan, Mandarake, Suruga-ya, Rakuten Ichiba, Amazon Japan, Yahoo! Shopping. Week 40 covers transactions from September 28 – October 4, 2026.

japan #marketdata #animefigures #collectibles #datasets #reselling #priceintelligence #apify

W39_STRIpped_TEST

atu ino — Sun, 04 Oct 2026 21:31:45 +0000

Weekly Market Pulse — Japan Hobby & Collectibles (Week 39, 2026)

Fresh weekly price data is now available for Japanese anime figures, plastic models (Gunpla), trading cards, dolls, and collectibles. This week's dataset covers 7 major Japanese marketplaces with 1,000+ price records from actual sold and listing data.

📊 What's in This Week's Dataset

Marketplace	Sample Items	Price Range (JPY)	Key Trends
Yahoo! Auctions Japan	400+	¥500 – ¥150,000	High-end figures trending upward
Mercari Japan	350+	¥300 – ¥80,000	Volume leader for mid-range items
Mandarake	200+	¥1,000 – ¥200,000	Rare/vintage premium holding strong
Suruga-ya	150+	¥800 – ¥120,000	Consistent retailer pricing
Rakuten Ichiba	100+	¥2,000 – ¥90,000	New release pre-orders active
Amazon Japan	80+	¥1,500 – ¥60,000	Prime-eligible fast movers
Yahoo! Shopping	60+	¥1,000 – ¥70,000	Bundle deals emerging

🎯 Top Movers This Week

Rising:

Gunpla (MG/RG/PG grades): +8–12% on limited re-releases
Vintage 1980s–90s figures: Steady collector demand
Trading card singles (holos/alt-arts): Volatile but upward bias

Stable/Declining:

Common prize figures (¥1,000–3,000): Oversupply, price compression
Standard edition Nendoroids: Flat, ample stock

📈 Why This Data Matters

Actual sold prices (not asking prices) from completed transactions
Condition-graded data (Mint/Opened/Box Damaged/etc.)
Brand, series, character, release date metadata for filtering
Historical trend columns for week-over-week comparison

🔗 Get the Data

Product	Price	Description	Link
Free Sample (30 rows, 7 marketplaces)	FREE	Frozen snapshot — try before you buy	https://atushi5.gumroad.com/l/kutuxe
Full Weekly Dataset (1,000+ rows, weekly updates)	$39.99	Complete CSV with all marketplaces, historical trends, condition grading	https://atushi5.gumroad.com/l/agyhq
Weekly Deep-Dive Market Report	$10	Analysis, charts, top movers, resale strategy notes	https://atushi5.gumroad.com/l/qdyyyi

🛠 Technical Details

Format: CSV (UTF-8), compatible with Excel, Python/pandas, Google Sheets, Power BI, Tableau
No API key required — direct download after purchase
Updated every Monday with the prior week's completed transactions
Historical archive included — build your own price models

📚 Use Cases

Resale arbitrage: Buy undervalued, sell at market rate
Collection valuation: Know what your shelf is worth
Market research: Track Japanese hobby demand cycles
AI/ML training: Clean, labeled price data for forecasting models
Import/export: Calibrate buy prices for Japanese sourcing

This summary is generated from the weekly Japanese Hobby & Collectibles Market Price Dataset. Data sources: Yahoo! Auctions Japan, Mercari Japan, Mandarake, Suruga-ya, Rakuten Ichiba, Amazon Japan, Yahoo! Shopping. Week 39 covers transactions from September 21–27, 2026.

japan #marketdata #animefigures #collectibles #datasets #reselling #priceintelligence

This Week's Japanese Hobby & Collectibles Market Price Summary (Week 39, 2026)

atu ino — Sun, 04 Oct 2026 21:31:06 +0000

simple test

W39_SIMPLE_TEST

atu ino — Sun, 04 Oct 2026 21:31:03 +0000

simple test

TEST_W39_UNIQUE_1791149425

atu ino — Sun, 04 Oct 2026 21:30:26 +0000

test unique

Japanese Market Data You Can Actually Use: 8 Apify Actors for Scraping Mercari, Yahoo Auctions, Rakuten, and More

atu ino — Sun, 04 Oct 2026 21:10:23 +0000

title: Japanese Market Data You Can Actually Use: 8 Apify Actors for Scraping Mercari, Yahoo Auctions, Rakuten, and More
tags: web-scraping, data, apify, japan, ecommerce
published: false

Japanese Market Data You Can Actually Use: 8 Apify Actors for Scraping Mercari, Yahoo Auctions, Rakuten, and More

If you're building a price comparison tool, a market research dashboard, or an AI agent that needs real Japanese ecommerce data, you've probably hit the same wall: Japan's biggest marketplaces don't have public APIs — or they're locked behind expensive enterprise contracts.

Mercari, Yahoo Auctions, Rakuten Market, SUUMO, Kakaku.com — these are the platforms where actual transactions happen. But getting structured data out of them requires either:

Reverse-engineering their internal APIs (fragile, often blocked)
Running a headless browser with the right session cookies
Using someone else's work

We've been doing #1 and #2 for over a year, and we've turned what we learned into 8 Apify Actors on the Apify Store.

What Each Actor Does

Actor	Target	What It Extracts
mercari-japan-search-scraper	Mercari Japan	Search results, prices, listing details, sold data
yahoo-auctions-japan-scraper	Yahoo Auctions	Active/closed listings, bid history, seller ratings
japan-kakaku-price-search	Kakaku.com	Product price comparisons across Japanese retailers
suumo-japan-real-estate-scraper	SUUMO	Rental/property listings with price per square meter
japan-market-mcp	Multi-platform	Unified MCP interface for Japanese market data
rakuten-japan-mcp	Rakuten Market	Product catalog + pricing via MCP
mercari-japan-scraper	Mercari Japan	Full listing scraper with image URLs
japan-prize-giveaway-scraper	Japanese X Giveaways	Campaign tracking for sweepstakes automation

Why This Matters for AI Agents

If you're building an agent that needs to answer questions like:

"What's the going rate for a used Sony α7 IV in Japan?"
"How do camera gear prices fluctuate across Mercari vs. Yahoo Auctions?"
"What's the rental market doing in Shibuya vs. Shinjuku?"

…you need real data. And Japanese market data is notoriously hard to get. These Actors handle the scraping complexity so you don't have to.

The Hard Parts We Solved

Mercari's search API is not callable directly. The search page is a fully client-side rendered Next.js app. The JSON endpoint exists (POST https://api.mercari.jp/v2/entities:search) but requires session tokens that change on every page load. Our Actor handles the token extraction and rotation automatically.

Yahoo Auctions rate limits. Yahoo is aggressive about bot detection. We use residential-proxy rotation + request pacing to stay under the radar while still hitting the endpoints that matter.

Kakaku.com's dynamic pricing. Kakaku.com renders prices client-side via JavaScript. Our Actor uses headless Chrome to extract the final rendered prices, not the raw HTML.

Get Started

Each Actor has:

Example run input — paste it in and hit "Run" to see results immediately
Dataset output — structured JSON/CSV you can query or pipe into your pipeline
Scheduled runs — set it and forget it, with daily/weekly refreshes

Try the mercari-japan-search-scraper first — it's the most mature and has the richest output schema.

Data Used in This Post

The market data behind our weekly Japanese ecommerce reports comes from these same Actors. If you find this useful, star the repo on GitHub — it helps other developers discover these tools.

Weekly Update: 650+ Anime Figure Prices Now Available Free on GitHub

atu ino — Sun, 04 Oct 2026 16:36:23 +0000

Weekly Update: 650+ Anime Figure Prices Now Available Free on GitHub

Every Monday, I push a fresh snapshot of Japanese anime figure prices to GitHub Releases.

What's in the data

655 figures, each with:

Current lowest/highest price across 5+ Japanese retailers
Stock availability (in/out of stock counts)
Shop-level offer details (price, URL, condition)
MSRP, scale, sculptor, release date

The raw CSV is ~900 KB — small enough to download in seconds, rich enough to build models on.

Why free?

I scrape this data weekly for my own resale research. Sharing it publicly serves two purposes:

Reproducibility — others can verify my price-tracking claims
Network effects — more users = more feedback = better data quality

The full historical dataset (price history over months, not just current snapshot) is available on Gumroad for researchers who need time-series analysis.

Technical details

Source: MyFigureList (primary) + secondary aggregators
Normalization: unified schema, confidence scoring, deduplication
Delivery: automated weekly GitHub Release via Python script + GitHub Actions
Last updated: {{DATE}}

How to use

# Download latest release asset
curl -L -o figure_prices.csv.gz \
  $(curl -s https://api.github.com/repos/atushi1841/kensho/releases/latest \
  | grep browser_download_url \
  | grep csv.gz \
  | cut -d\" -f4)

# Or use the API
curl -s https://api.github.com/repos/atushi1841/kensho/releases/latest \
  | jq '.assets[].browser_download_url'

This is part of the Kensho project — automated Japanese marketplace price tracking.

Japanese Market Data You Can Actually Use: 8 Apify Actors for Scraping Mercari, Yahoo Auctions, Rakuten, and More

atu ino — Sun, 04 Oct 2026 14:58:08 +0000

Mercari, Yahoo Auctions, Rakuten Market, SUUMO, Kakaku.com — these are the platforms where actual transactions happen. But getting structured data out of them requires either:

Reverse-engineering their internal APIs (fragile, often blocked)
Running a headless browser with the right session cookies
Using someone else's work

We've been doing #1 and #2 for over a year, and we've turned what we learned into 8 Apify Actors on the Apify Store.

What Each Actor Does

Actor	Target	What It Extracts
mercari-japan-search-scraper	Mercari Japan	Search results, prices, listing details, sold data
yahoo-auctions-japan-scraper	Yahoo Auctions	Active/closed listings, bid history, seller ratings
japan-kakaku-price-search	Kakaku.com	Product price comparisons across Japanese retailers
suumo-japan-real-estate-scraper	SUUMO	Rental/property listings with price per square meter
japan-market-mcp	Multi-platform	Unified MCP interface for Japanese market data
rakuten-japan-mcp	Rakuten Market	Product catalog + pricing via MCP
mercari-japan-scraper	Mercari Japan	Full listing scraper with image URLs
japan-prize-giveaway-scraper	Japanese X Giveaways	Campaign tracking for sweepstakes automation

Why This Matters for AI Agents

If you're building an agent that needs to answer questions like:

"What's the going rate for a used Sony α7 IV in Japan?"
"How do camera gear prices fluctuate across Mercari vs. Yahoo Auctions?"
"What's the rental market doing in Shibuya vs. Shinjuku?"

…you need real data. And Japanese market data is notoriously hard to get. These Actors handle the scraping complexity so you don't have to.

The Hard Parts We Solved

Yahoo Auctions rate limits. Yahoo is aggressive about bot detection. We use residential-proxy rotation + request pacing to stay under the radar while still hitting the endpoints that matter.

Kakaku.com's dynamic pricing. Kakaku.com renders prices client-side via JavaScript. Our Actor uses headless Chrome to extract the final rendered prices, not the raw HTML.

Get Started

Each Actor has:

Example run input — paste it in and hit "Run" to see results immediately
Dataset output — structured JSON/CSV you can query or pipe into your pipeline
Scheduled runs — set it and forget it, with daily/weekly refreshes

Try the mercari-japan-search-scraper first — it's the most mature and has the richest output schema.

Data Used in This Post

The market data behind our weekly Japanese ecommerce reports comes from these same Actors. If you find this useful, star the repo on GitHub — it helps other developers discover these tools.

懸賞応募自動化のリアル: 3万円の収益を生んだコード

atu ino — Sun, 04 Oct 2026 02:03:29 +0000

懸賞応募自動化のリアル

成果

8アカウントで毎日130-200件の懸賞に応募し、月1-3万円の収益を達成。

技術的戦略

Playwright + Chrome CDPでブラウザ操作を自動化
プロキシ分離で垢ごとに独立IP
ランダム遅延で人間風の動作パターン
曜日制御で同一時間帯の重複を回避

BOT検出対策

マウス移動・スクロールのシミュレーション
1日20アクション以下の抑止
アカウントごとの行動パターンの多様化
スパム行為の完全排除

収益のボトルネック

「作る」より「売る」。Apify Storeやdev.toでの可視性向上に注力中。

Published on dev.to by @atushi1841

Data used in this post

The datasets behind this analysis are available on Apify (pay-per-result, free tier to start):

japan-prize-giveaway-scraper

MCPサーバー6本をMCP公式レジストリに登録した結果

atu ino — Sun, 04 Oct 2026 02:03:23 +0000

MCPサーバー6本をMCP公式レジストリに登録した結果

Model Context Protocolとは

Claude DesktopやCursor、WindsurfなどのAIツールが外部データにアクセスするための標準プロトコル。2025年に安定版がリリースされ、急速に普及中。

公開したMCPサーバー

japan-minimum-wage-mcp - 日本全国の時給データ
japan-jepx-mcp - 日本電力取引所の需給予測
japan-rent-market-mcp - 家賃相場データ
japan-property-hazard-mcp - 住宅危険因子データ
kensho-sweep-mcp - 懸賞応募データ
japan-prime-product-mcp - Amazonプライム商品価格

登録したレジストリ

mcp.so（中国最大のMCPレジストリ）
GLaMA（AIコミュニティ）
PulseMCP
Cursor Marketplace
GitHub (public repo)

効果測定

登録後72時間での外部的利用は確認できていません。可視性向上は第一歩。

Published on GitHub & MCP Registries

Apifyで日本市場の価格データを自動収集: 86個のアクターを公開

atu ino — Sun, 04 Oct 2026 02:02:13 +0000

Apifyで日本市場の価格データを自動収集

プロジェクト概要

86個のApifyアクターを公開し、日本のEC・オークション・不動産・グルメデータをAPI経由で取得できるようにしました。

対象データソース

Mercari Japan（メルカリ）
Yahoo!オークション
楽天市場
SUUMO / At Home（不動産）
食べログ / ぐるなび
Kakaku.com（価格.com）
デジマート（楽器）

アクセス方法

Apify Store: https://apify.com/fruitful_quintessence

すべてPPE（Pay Per Execution）モデル。無料枠からテスト可能。

MCPサーバー対応

6つのMCPサーバーも公開中:

japan-minimum-wage-mcp
japan-jepx-mcp
japan-rent-market-mcp
japan-property-hazard-mcp
kensho-sweep-mcp
japan-prime-product-mcp

技術スタック

Python + Playwright（headless）、プロキシ分離、自動レポート生成。詳細はGitHubを参照。

Published on Apify & GitHub by @atushi1841

懸賞112件の自動応募ログを全部集計したら、「応募導線」と「当選枠」に地味な崖があった

atu ino — Tue, 29 Sep 2026 00:01:04 +0000

title: 懸賞112件の自動応募ログを全部集計したら、「応募導線」と「当選枠」に地味な崖があった
tags:

python
automation
datajournalism
japanese private: true

懸賞112件の自動応募ログを全部集計したら、「応募導線」と「当選枠」に地味な崖があった

国内懸賞 112件を自動収集し、8アカウントで 120件のエントリーを実行した。その生ログから、応募導線・当選枠・締切の分布を機械可読な形で集計したのがこのレポートだ。手作業では絶対に作れない「112件横断の傾向」を見ていく。

何を測ったのか

対象: data/collected_today.json ほかスナップショット（112件）
実行: 自動応募 120件（アカウントあたり案件カバー率 13.4%）
判明した当選枠合計 2,286口 / 推定賞品価値 47,030円

わかったこと(1): 伸びているキーワード

最も多く出現したのは #プレゼント企画（15件）、BOOTH（7件）、#その場で当たる（6件）、Amazon（6件）、#VRChat（6件）、菓子（5件）、チケット（5件）、毎日（4件）。単なる出現数ではなく、「そのキーワードの案件にどれだけ自動応募が刺さったか（応募率リフト）」も併記している。リフトが正のキーワードは、運用側の判断と需要が噛み合っている語だ。

わかったこと(2): 当選枠の構造

枠数構成比は 2〜5名 37件、1名（最難関） 26件、枠数不明・記載なし 23件の順。枠が大きい案件ほど当選確率は上がるが、実際に当選が積み上がっている枠レンジは実測当選テーブルを見ればわかる。

わかったこと(3): 参加導線という見えない壁

応募導線は X 98.2%、LINE 0.9%、未判定 0.9%。X内で完結する案件は自動化できるが、LINE・外部フォーム・会員IDは1件ごとに人手または専用実装が必要で、ここが実質的な参入障壁になる。

再現方法

python3 scripts/kensho_data_journalism.py --week 2026W39

統計JSONは reports/journalism/2026W39.json に出力される。同じ入力からは同じ数値が出る（乱数も時刻依存の集計も使っていない）。

まとめ

自動化の勝率を上げるのに効くのは、根性ではなく「構造の可視化」だった。
どの導線が自動化できて、どの枠レンジに当選が積み上がり、どのキーワードの案件に応募が回っていないか——
それらを毎週機械的に出しておくと、改善の順番が迷わなくなる。

対象データ: 112件 / 応募 120件 / 統計ハッシュ 24996894bc04c75d（qiita 用下書き・自動生成）

Data used in this post

The datasets behind this analysis are available on Apify (pay-per-result, free tier to start):

japan-prize-giveaway-scraper