FlashScore News Scraper

What does FlashScore News Scraper do?

FlashScore News Scraper pulls the news list, the news category/tag menu, and individual article bodies from FlashScore. It talks directly to the same persisted-query GraphQL API the official FlashScore Android app uses (global.ds.lsapp.eu), reverse-engineered by live MITM capture against a real device (Frida SSL-pinning bypass + mitmproxy). No FlashScore account, no API key, no scraping setup.

This is the news slice of FlashScore All-in-One API, split out as its own focused, batch-run Actor.

Why use FlashScore News Scraper?

• Real news feed, not scraped HTML — the exact persisted-query GraphQL calls the app itself sends, cross-checked live
• Discover real tag ids — includeMenu returns every category/team/tournament tag FlashScore's news section supports (dozens of sports, hundreds of teams/tournaments), which you can then pass as tagId to filter the news list
• Article bodies in parallel — pass a batch of article ids and get them all fetched concurrently, one row per article
• No account needed — every read request works fully anonymously
• Use cases: sports news aggregation, content monitoring, editorial research

Input

Field	Type	Description
fetchNewsList	boolean	Fetches the news list (optionally filtered by tagId). Default true.
tagId	string	Filters the news list to one tag (team, league, sport, or topic). Sourced from a tag's id field in the news menu output — there's no fixed short list, FlashScore has hundreds of tags.
includeMenu	boolean	Fetches the news category/tag menu. Default false. Useful to discover real tagId values.
articleIds	string[]	Article ids to fetch in full, in parallel. Default []. Often discoverable from the news list output's nested article stub id fields.

At least one of fetchNewsList, includeMenu, or a non-empty articleIds must be set, or the run throws immediately.

Example:

{
    "fetchNewsList": true,
    "includeMenu": true,
    "articleIds": ["QFgdUkIk"]
}

Output

The parent Actor's listNews/getNewsMenu were never typed beyond unknown — their real response shapes hadn't been verified. This Actor's live smoke test found and confirmed the following real shapes.

News list (fetchNewsList: true) — the response is data.findNewsLayoutByProjectId.sections, an array of named sections (e.g. "Top News", "Transfer News", "Premier League"), each carrying a small articles array of stubs (id + isTopped only — not full article bodies). There is no single flat array of individual articles upstream; sections are the real top-level list. One dataset row is pushed per section, tagged type: "news-item":

{
    "type": "news-item",
    "__typename": "CatSection",
    "title": "Transfer News",
    "articles": [
        { "__typename": "CatArticleOnLayoutSection", "id": "v5bCE1WG", "isTopped": true },
        { "__typename": "CatArticleOnLayoutSection", "id": "MZIltPuS", "isTopped": false }
    ],
    "sectionType": "MIXED",
    "tag": {
        "__typename": "CatTag",
        "id": "pG5roijlWSzc94ws",
        "entityId": "pG5roijl",
        "entityType": "TAG",
        "slug": "transfer-window"
    }
}

Use each section's tag.id as tagId to filter, and each article stub's id as an articleIds entry to fetch full content.

News menu (includeMenu: true) — the response is data.findNewsMenuByProjectId.items, a clean flat array of tags. One row per tag, tagged type: "news-tag":

{
    "type": "news-tag",
    "__typename": "CatMenuItem",
    "name": "Premier League",
    "tag": {
        "__typename": "CatTag",
        "id": "dYlOSQODCdnS0XT8",
        "entityId": "dYlOSQOD",
        "entityType": "TOURNAMENT_TEMPLATE",
        "slug": "premier-league"
    }
}

Article (articleIds) — one row per id, tagged type: "article":

{
    "type": "article",
    "articleId": "QFgdUkIk",
    "article": {
        "data": {
            "findCatArticleById": {
                "id": "QFgdUkIk",
                "type": "STANDARD",
                "publishedAt": 1788130743,
                "title": "Premier League talking points: Chelsea dazzle in attack as Liverpool stutter",
                "slug": "football-premier-league-premier-league-talking-points-chelsea-dazzle-in-attack-as-liverpool-stutter",
                "images": [{ "url": "https://livesport-ott-images.ssl.cdn.cra.cz/...jpeg", "altText": "..." }]
            }
        }
    }
}

If any of the three sections above ever return an unrecognized shape (upstream schema change), this Actor falls back to pushing the whole raw GraphQL result as one row (type: "news-list" or type: "news-menu") instead of guessing a wrong structure.

Known limitations

• The parent Actor's own types for listNews/getNewsMenu were unknown — never previously verified. This README documents the real shapes found during this Actor's live smoke test (above); a future upstream schema change could alter them.
• The getArticle persisted query (nah) returns article metadata (title, slug, publish date, images) but not a body/content field in the observed live response — the persisted query's field selection is fixed server-side and this Actor can't request additional fields.
• "News items" in the list response are really named sections (Top News, Transfer News, per-league sections), each holding article stubs (id only) rather than a single flat list of full articles — fetch full content via articleIds.
• This Actor only covers /news, /news/menu, and /news/{id}. Matches, standings, search, and everything else FlashScore-related live in the parent FlashScore All-in-One API.

Pricing

Pay-per-event. Charged $0.004 per news-item event (one event per dataset row pushed — news list section, menu tag, or article). See the Actor's Pricing tab for current tiered rates.