Discord Public Server Search & Watchlist

Find relevant public communities by keyword/category and refresh selected server URLs for community market research. Export community names, descriptions, aggregate members/online counts, source locale and configuration flags. URL watchlist rows can also include visible categories, tags, join link, boosts and creation time. This unofficial Actor uses the official anonymous directory HTML.

Input

Use queries, optional category, and/or serverUrls. Up to5 query/category discovery sources and50 direct public server URLs. Combined inputs must satisfy a preflight workload estimate of1 robots request + direct URL count + discovery source count × ceil(maxServersPerSource/12) <=90; reduce the per-source limit or split runs otherwise. Five queries with50 direct URLs work at the default50 per-source ceiling (76 estimated requests); five queries at200/source permit up to4 direct URLs (90 estimated requests). Network delays and rate limits can still exhaust the300-second deadline before delivery. Empty input defaults to queries=["python"]; when any source field is supplied, omitted source fields are empty. Category-only input browses that category. maxServersPerSource defaults50 with200 ceiling; maxItems defaults500 with1000 ceiling. Category options are Gaming, Entertainment, Education, Music, Science & Tech. URL watchlists accept only exact discord.com/servers/SLUG-ID detail URLs.

Output

One serverId produces one row with screened name/description, sourceUrl, approximate memberCount/onlineCount, sourceInputs, sourceRank, coverage and observedAt. Search rows also preserve visible locale, verified/partnered flags and features where provided. Direct detail rows add only visible categories, tags, public inviteUrl, boostCount and createdAt. Optional unavailable fields are omitted. Source search metadata can be older than detail-page counters; observations are snapshots, not a verified growth rate. Preserve source rank and do not sort silently. SUMMARY records request/page counts, source estimated totals, distinct rows, filtering and cap/actual-next-link stops; it is uncharged in key-value storage. Keyword queries return the currently accessible anonymous SSR search window, which can end below the source estimate while the HTML still advertises later offsets. An empty window stops only when both JSON-LD and source search hit state explicitly contain no records; SUMMARY marks publicWindowEnded, continuationAdvertised, last attempted/nonempty offsets and next advertised URL. This is a bounded window boundary, never completion of all indexed communities. Result totals are estimates and do not promise complete coverage. Category browsing follows its separately available pages.

Charging

$0.00075 per unique delivered server through server-exported; $0.75/1000. No actor-specific start event or trial. Required sources are fully collected into a bounded sanitized buffer before any delivery/billing. A source failure leaves zero rows/events. Duplicate matches merge sourceInputs and are delivered/charged once. Filtered records, metadata and SUMMARY are uncharged. A run with no delivered servers fails without charging; mixed evidenced empty queries appear in SUMMARY when another source returns rows.

Limits

256MB, HTTP/HTML only; no login, token, API credentials, paid API, proxy, browser or media downloads. Search pages typically12 rows; follow actual rel=next offset links and require strictly increasing offset plus distinct progress. Short pages alone are not exhaustion. Reject changed/duplicate-only pages, unsafe next URLs, missing counts and blocked/login responses. At most90 requests,15-second request timeout,3-second spacing (at most20 requests/minute), up to2 bounded retries for429 with a numeric Retry-After of at most30 seconds,2MiB response cap and300-second whole-run timeout. Requests stay in discord.com/servers; public invite links are returned as metadata and never visited. Rich search fields use a bounded static-literal lexer on source SSR payloads; never execute source JavaScript. Direct watchlist supports bulk known communities, not personal or private server membership.

Privacy

Collect aggregate community metadata only. Omit all user/member identifiers, account profiles, channels, messages, contacts, credentials, precise locations and media URLs/files. Public directory community IDs/names and join links are kept only for community attribution. Contact patterns, @mentions, unsafe links and secret-like strings are redacted from name/description/tags before persistence. A server may be associated with a person and pattern screening does not guarantee full anonymization. No automated joining, messaging, personal enrichment or inferred user characteristics.

Source and terms

Discord robots currently permits public /servers HTML; prohibited /channels, /profile and general /api paths are never requested. Recheck robots every run and honor Disallow. Current terms restrict scraping without written consent and commercialization of source data. Exact applicable excerpts and DR-0007 treatment are recorded in the scout with strict public/aggregate guards. Free alternatives include manual directory browsing and maintaining your own client; paid value is managed bounded discovery/watchlist exports and scheduling, not privileged access. Independent creation review and private cloud default/nondefault/max-load verification remain release gates.

If something goes wrong

• Wrong number or a failed run? Open a ticket on the Issues tab. I read every one and reply within 2 business days (Japan time).
• You never get a fake "empty" result. If the site can't be read, the run fails and says so.
• Delivered rows only. One validated unique row is one billable event. Duplicates, summaries, failed acquisition and undelivered rows are not charged.
• Checked every week. An automatic test runs this tool weekly; if the site changes, I fix it.
• Public source data only. No login or member/applicant records. Contact patterns are screened before delivery; prose may still name people.

More tools by the same author

• Mercari Japan Sold Prices — What Items Really Sell For
• Yahoo! Auctions Japan Sold Prices — Median, Range, Bids

All tools (Japan marketplaces, real estate, jobs, racing, prediction markets): https://apify.com/jpmarketdata

Disclaimer

Unofficial, independent tool — not affiliated with, endorsed by, or sponsored by Discord. Product names and logos belong to their owners and only say where the data comes from. Data is read from public pages, for market research; check before you act on it.