🎮 Steam Scraper: Reviews, Achievements, News & Players

Steam Scraper pulls game details, user reviews with language and sentiment filters, achievement unlock rates, live player counts and news or patch notes for any Steam game, from an App ID, a store URL, a game name, a keyword or a tag. It reads Valve's own public Store and Web APIs, so it needs no login, no browser and no API key. Add the only-new mode to get just the reviews and news published since your last run. Export as CSV, Excel, JSON or XML, or call it by API.

Steam Scraper turns a Steam App ID like 730 (Counter-Strike 2), a store URL or a game name like Stardew Valley into structured rows. You can also discover games by keyword, by tag (Roguelike) and by filters such as minMetacritic and maxPrice. One run can return the store record, up to 10,000 reviews per game, achievement unlock percentages, a live player snapshot and up to 200 news items. Each row carries a recordType field, so you can filter one dataset into clean tables.

📋 What does Steam Scraper do?

Steam Scraper collects five public data types from Valve's endpoints and one optional extra:

• Game details: name, price, currency, discount, release date, developers, genres, top 20 store tags, platforms, Steam Deck status, Metacritic score, DLC IDs, system requirements and trailer links. Add priceCountries for a regionalPrices list (up to 20 countries).
• Review summary on the game row: reviewScoreDesc (for example Overwhelmingly Positive), totalPositive, totalNegative, totalReviews and positivePercent.
• Reviews: text, thumbs up or down, helpful votes, playtime, language and purchase flags. Filter by language, sentiment, order and purchase type, and fetch up to 10,000 per game.
• Achievement stats: the global unlock percentage of every achievement.
• Live player count: one concurrent-players snapshot per game. Schedule runs to build a history.
• News: announcements and patch notes, up to 200 per game.
• Achievement schema (optional): full names, descriptions and icons. Needs your own free Steam Web API key.
• Discovery: searchQueries and searchTags find games, and minMetacritic, onSaleOnly, freeOnly and maxPrice filter them.
• Only new: with onlyNew, reviews and news already delivered under the same stateKey are skipped and never billed.

Prices default to USD (country is us) and English; pick another store in country.

📊 What data can I extract from Steam?

The dataset schema has 92 fields and six views.

recordType	Main fields	Notes
game-record	name, price, priceCurrency, discountPercent, genres, tags, metacriticScore, reviewScoreDesc, totalPositive, regionalPrices	One row per game. Always on.
review-record	reviewId, votedUp, votesUp, playtimeForeverHours, language, review, steamPurchase, timestampCreated	On by default, 100 per game, up to 10,000.
achievement-stats-record	achievementName, globalUnlockPercent	Opt in. Names are developer-defined, some are codes like a0.
player-count-record	currentPlayers	Opt in. One live snapshot per game.
news-record	newsId, title, url, contents, publishedAt	Opt in. Up to 200 per game.
achievement-schema-record	achievementName, description, hidden, imageUrl	Needs your own Steam key.

🧪 What does the Steam Scraper output look like?

A game-record row for Stardew Valley (App ID 413150), trimmed to a few columns. Rows for other record types carry only the fields that apply to them:

{
  "recordType": "game-record",
  "appId": 413150,
  "name": "Stardew Valley",
  "price": 14.99,
  "priceCurrency": "USD",
  "genres": "Indie, RPG, Simulation",
  "metacriticScore": 89,
  "reviewScoreDesc": "Overwhelmingly Positive",
  "totalPositive": 885208,
  "totalNegative": 13258
}

An illustrative review-record, a player-count-record and an achievement-stats-record (values shortened):

{ "recordType": "review-record", "appId": 413150, "votedUp": "No", "votesUp": 12, "language": "english", "playtimeForeverHours": 3, "review": "Short review text..." }
{ "recordType": "player-count-record", "appId": 413150, "gameName": "Stardew Valley", "currentPlayers": 54008 }
{ "recordType": "achievement-stats-record", "appId": 413150, "gameName": "Stardew Valley", "achievementName": "a0", "globalUnlockPercent": "62.7" }

When Valve has no value for a field, the field is "N/A", so you can tell "missing at the source" from a bug.

💰 How much does it cost to scrape Steam?

The Actor uses pay-per-event pricing. You pay only for rows that are saved to your dataset. Error rows, "game not found" rows, empty results and rows skipped by the only-new mode are never billed. Prices at the Free plan tier (paid plans get lower rates, down to $0.0102 per game record at Gold and above):

Event	Price per row	Per 1,000 rows
game-record	$0.012	$12.00
review-record	$0.01	$10.00
achievement-stats-record	$0.01	$10.00
player-count-record	$0.01	$10.00
news-record	$0.01	$10.00
achievement-schema-record	$0.01	$10.00

Worked examples at the Free tier: one game with 250 English negative reviews is 251 rows, about $2.51 ($0.012 + 250 x $0.01). Counter-Strike 2 with 300 reviews, achievement stats, a player count and 200 news items was 503 rows, about $5.01. A single game record with reviews turned off costs $0.012. Regional prices, tags and the review summary add no charge. Discovered games are billed as game-record and count toward maxItems. Set Max Items or a maximum cost per run and the Actor stops cleanly. Free users are limited to 10 items per run.

🚀 How to scrape Steam in 3 steps

1. Open the Actor and enter Steam App IDs (or store URLs), game names, a search keyword or a tag. The form pre-fills App ID 730: clear it for a search only.
2. Switch on the extras you need and set Max Items.
3. Click Start, then download the dataset as CSV, Excel, JSON or XML.

⚙️ Input

Field	Type	What it does
appIds	list of strings	Numeric Steam App IDs or store URLs, for example 730.
gameNames	list of strings	Names resolved to an App ID (top match).
searchQueries	list of strings	Store keyword search, each term searched separately.
searchTags	list of strings	Tag or genre names or IDs. A game must match all of them.
maxGamesPerSearch	integer	Games per search. Default 10, cap 500. Order with searchSort.
minMetacritic, maxPrice	integer, number	Skip games below the score or above the price.
onSaleOnly, freeOnly	boolean	Only discounted or only free games. Cannot be combined.
maxItems	integer	Total rows across all record types. Free users: 10.
includeReviews	boolean	Individual user reviews. Default on.
reviewLanguage, reviewType	string	Review language (30 plus all) and all, positive or negative.
reviewFilter, reviewDayRange	string, integer	Order: recent, updated, all (most helpful, 1 to 365 day window).
reviewPurchaseType, maxReviewsPerGame	string, integer	steam (default), all or non_steam_purchase; reviews per game, default 100, up to 10,000.
includeAchievementStats, includePlayerCount	boolean	Unlock percentages; live concurrent players.
includeNews, maxNewsPerGame	boolean, integer	News and patch notes, default 20, up to 200.
country, priceCountries	string, list	Store country, and up to 20 countries for regionalPrices.
onlyNew, stateKey, resetSeen	boolean, string, boolean	Only-new monitor under a state label.
steamApiKey	secret string	Optional. Your own free key adds the achievement schema.

Give at least one App ID, game name, search keyword or tag; otherwise the run ends with a free error row.

Example input, negative English reviews of Stardew Valley:

{
  "appIds": ["413150"],
  "reviewLanguage": "english",
  "reviewType": "negative",
  "maxReviewsPerGame": 250,
  "onlyNew": true,
  "stateKey": "stardew-watch"
}

📤 Output

Every row goes to the default dataset and has recordType and appId. If an App ID is not found you get one free row with an error message and the run continues. With onlyNew on, the game row, achievement stats and player count are delivered at most once per day and game, and reviews and news already delivered are skipped before billing. Reviews come newest first, so a capped first run never backfills older reviews later.

📈 How does it compare with other Steam scrapers on Apify?

Store pages read on 2026-10-10. A cell says "not shown" when the rival's page does not mention the feature.

Feature	This Actor	eiv/steam-scraper	cryptosignals/steam-scraper
Achievements and news	Unlock rates, schema, news	Not shown	Not shown
Review controls	Language, type, order, 10,000	Language and order	Language and filter, 500 max
Discovery	Keyword, tag, score, price	Keyword, genre, price, score	Keyword search
Only-new mode	Yes	Not shown	Not shown
Regional prices	Up to 20 countries	Yes, per country	Not shown
Price per 1,000	$12 game, $10 review	$4 game, $0.40 review	$10 result

Rivals are cheaper per row. eiv/steam-scraper charges $0.004 per game and $0.0004 per review plus a $0.005 start, so 1,000 reviews cost $0.40 there against $10.00 here, 25 times more. constant_quadruped/steam-review-monitor ($0.004 per item) also has a sinceLastRun mode with theme tagging, which we do not. Choose this Actor for breadth and the only-new mode, and a review-only actor for raw review volume.

🎯 Steam scraper use cases

Market research on prices and Metacritic scores, player-count tracking, sentiment and review monitoring with onlyNew, and game discovery by tag and price.

🔌 How do I use Steam Scraper through the API and integrations?

Call it with the Apify API or client libraries:

curl -X POST "https://api.apify.com/v2/acts/RecordsData~steam-games-player-stats/run-sync-get-dataset-items?token=YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"appIds":["730"],"includePlayerCount":true,"maxItems":10}'

It also works with Zapier, Make, n8n, Google Sheets and Airbyte, and through MCP (next section).

🤖 Steam Scraper MCP server for AI agents (Claude, Cursor, VS Code)

Connect this actor to any MCP client through the Apify MCP server at https://mcp.apify.com?tools=RecordsData/steam-games-player-stats. Config for clients that take a JSON MCP config:

{
  "mcpServers": {
    "apify": {
      "url": "https://mcp.apify.com?tools=RecordsData/steam-games-player-stats"
    }
  }
}

Example agent prompt: "Use the Steam Scraper for Stardew Valley and Counter-Strike 2 with live player count and news on, reviews off, then tell me the current price, discount, Metacritic score and how many players are online right now. Then find Roguelike games under 15 dollars with Metacritic 80 or more."

An agent should choose this tool when it needs Steam store details, review totals, user reviews filtered by language or sentiment, achievement unlock rates, live concurrent players, patch notes, or a list of games found by keyword, tag or price, given an App ID, a name or a search.

⚖️ Is it legal to scrape Steam data?

Steam Scraper reads public endpoints that Valve publishes. It collects only public game, review, achievement and news data. Reviews contain public usernames, so handle them under the privacy rules that apply to you. You are responsible for using the data in line with Valve's terms. This Actor is independent and not affiliated with or endorsed by Valve Corporation or Steam.

❓ Frequently asked questions

Do I need a Steam API key to use Steam Scraper? No. Steam Scraper needs no key for game details, reviews, achievement stats, live player counts or news, because it reads Valve's public Store and Web endpoints. Only the optional achievement schema (full names, descriptions and icons) needs your own free Steam Web API key from steamcommunity.com/dev/apikey, entered in the steamApiKey field.

Can I search Steam by game name instead of App ID? Yes. Enter a name in the gameNames field and Steam Scraper resolves it to an App ID through Steam search, for example "Stardew Valley" resolves to App ID 413150. The top match is used, so prefer a numeric App ID when a name is ambiguous. For several games at once, use searchQueries or searchTags.

What happens if I send an empty or invalid input to Steam Scraper? The run does not crash. Input mistakes (no App IDs, names, keywords or tags, freeOnly together with onSaleOnly, or an unknown tag) end the Steam Scraper run as SUCCEEDED with one free error row (errorType: invalid-input, plus error and hint fields) and nothing is charged.

Why did Steam Scraper return 0 results? The App ID or name did not match a game, the game was removed from Steam, or your filters (minMetacritic, maxPrice, onSaleOnly) skipped it. An unknown ID gets a free error row and the run continues with the other IDs. If the run finds nothing at all, it still ends with a clear status message and charges nothing.

Why are some Steam achievement names codes like a0? Valve's free achievement endpoint returns the internal name the developer set, and Steam Scraper passes it through unchanged, so some games show readable names and others show codes such as a0. Add your own Steam key in steamApiKey to also get readable names and icons in achievement-schema-record rows.

How many Steam reviews do I get per game? By default Steam Scraper returns the 100 newest reviews per game in all languages. Set maxReviewsPerGame up to 10,000 and narrow them with reviewLanguage, reviewType (positive or negative), reviewFilter and reviewPurchaseType. At $0.01 per review row on the Free plan tier, 100 reviews for one game cost $1.00.

How fresh is the Steam player count? The player count is live at the moment of the run. One concurrent-player snapshot per game is returned in a player-count-record row. A single run does not return past values, so schedule the actor hourly or daily to build a history and chart launches, sales or patches over time.

Am I charged for failed or empty Steam Scraper rows? No. Steam Scraper bills only saved rows with a valid payload. Error rows, "game not found" rows and empty results are free, and empty or duplicate reviews are skipped before billing. With onlyNew, reviews and news you already received are skipped for free too, and you can set a maximum cost per run.

What currency are Steam Scraper prices in? Prices are in USD by default, from the US store (country is us) in English, so results do not depend on where the run starts. Change country for another store, or add priceCountries for a regionalPrices list. The priceCurrency field states the currency.

Why does my free Steam Scraper run stop at 10 items? Free Apify plan runs are limited to 10 items in total across all record types, so a game row plus nine reviews already reaches the limit. Paid plans can set any maxItems. At Free tier prices, a full 10-row preview costs about $0.10.

🔗 Want more game data?

• Steam Charts Scraper
• Steam Reviews Scraper
• GOG Scraper

🆘 Support

Questions or a custom field you need? Open an issue on the Actor page or write to contact.punkrecordsdata@gmail.com.

Last updated: 2026-10-11