Discord Scraper

Turn Discord's server directory and any invite link into a table: every public server with its member and online counts, boosts, category, tags, language, creation date, invite link, icon and banner. 100 servers in 14 seconds, no Discord account needed.

How it works

1. You type a keyword, pick a category, or paste links. A keyword (minecraft, anime, crypto) searches Discord's own directory of 52,000 discoverable servers, largest first. A category (Gaming, Music, Entertainment, Education) lists it or narrows the search. A server page, a directory search URL, an invite link (discord.gg/python) or a bare invite code reaches one server, listed in the directory or not.
2. Each server is read from Discord itself. Member and online counts as the directory shows them at run time, boosts, primary category and all categories, tags, enabled features, language, the vanity invite, icon, banner and splash. Switch on the server page and the row also carries the long description, the reasons to join, the social links and the full feature list.
3. One row comes back per server. 35 fields, flat, with null where Discord has nothing: numbers as numbers, dates in RFC 3339, a stable id and url, the creation date computed from the id, and how the row was found (foundBy, keyword, position, totalResults). Ready for a spreadsheet, a database or an AI pipeline.

✨ Why teams choose this over other Discord scrapers

Clicking through the directory 12 servers at a time to build a partnership list? Paying a fee per run start for a scraper that reads a third-party index with last month's counts? Getting back raw nested blobs with 29 translations per category and no creation date?

• 🔢 28 fields filled on every row, not 22 keys of raw output. Measured on the default run of 100 Minecraft servers: 28 of the 35 fields carry a value on every row, among them memberCount, onlineCount, boostCount, locale, createdAt, inviteUrl, category, categories, keywords, features and five flat flags such as isVerified and isPartnered. The most used directory scraper returns the raw directory hit: 22 keys, several of them search-highlighting noise and translation tables, no creation date, no flat flags.
• 🔀 Four ways in, one run. Keywords, a category browse, a server page or directory search URL, and an invite link or code. The most used directory scraper takes keywords and a category; none of the three most used ones resolves an invite.
• 🌍 Every public server, not only the directory. The directory lists 52,816 discoverable servers (23 Sep 2026). An invite link reaches any public server beyond it, with fields the directory does not show: verification level, NSFW level and boost tier. The most used scrapers read the directory, or a third-party index, only.
• 🏷️ A lower price per server than the most used Discord server scraper, and nothing per run start. Two of the three most used scrapers add a fee every time a run starts. Here a run that finds nothing bills nothing. The exact figure is on the Monetization tab of this page.
• ⚡ 100 servers in 14 seconds, 30 in 5. No browser loads a page: the run reads the data Discord's own pages carry and moves on. A whole category of 500 servers takes about a minute and a half.
• 📅 Server age on every row. createdAt comes from the server id itself, so every row says when the server was created, with no extra request. A young server with 100,000 members and a ten-year-old one read differently in an outreach list.
• 🧹 Clean rows an integrator can trust. memberCount is a number, not "1.2M"; createdAt and scrapedAt are RFC 3339 in UTC; id is the server id and url its permanent directory page; a field Discord does not have is null, never an empty string. Field names follow the ones the market already uses: memberCount, onlineCount, boostCount, inviteUrl, vanityUrl, iconUrl, bannerUrl, features, keywords.
• 🔓 No account, no token, no login. Type a keyword and run. Your own Discord account is never involved.

🎯 What you can do with Discord data

Team	What they build
Partnerships	The 200 largest gaming or anime servers with their online counts and invite links, ranked for sponsorship outreach
Game and product marketing	A weekly scheduled run over the servers around a title, with member and online counts compared week to week
Community management	A map of a category by size, language and creation date, to find where a community is missing
Web3 and crypto research	Every server matching a token, exchange or protocol name with its size, activity and age, to separate live communities from dead ones
Influencer and creator agencies	Servers of the Content Creator and Fandom categories with member counts and social links, to match creators with sponsors
Data and AI	A dataset of thousands of servers with descriptions, tags, categories, features and languages, for classification or research
Sales	A list of invite links from a CRM enriched with server names, ids, counts, verification levels and creation dates
Trust and safety research	NSFW flags, verification levels and boost tiers of a set of servers resolved from their invites

📥 Input parameters

Essentials

Parameter	Type	Description
keywords	array	Words to search in the directory: minecraft, anime, crypto, study. Matched against names, descriptions and tags, largest servers first. Up to 3,000 servers per keyword (the directory's limit).
category	string	gaming, music, entertainment, education, or empty for all categories. Narrows a keyword search, or lists the category when there is no keyword and no link.
startUrls	array	Server pages (https://discord.com/servers/midjourney-662267976984297473), directory search pages (https://discord.com/servers/gaming?query=rpg), invite links (https://discord.gg/python, discord.com/invite/xyz) or bare invite and vanity codes (python). Objects {"url": ...} work too.

🔢 Limits

Parameter	Type	Description
maxItems	integer	Servers to deliver in total across keywords, categories and links. Default 100. No fixed limit.

📄 Output

Parameter	Type	Description
includeDetails	boolean	Read each server's own page for about, reasonsToJoin, socialLinks and the full features list. One page per server, about two seconds each. Default false.

⚙️ Advanced

Parameter	Type	Description
debugMode	boolean	Detailed logs of every request and connection change. Default false.

With no keyword, no category and no link, the run lists the whole directory, largest servers first.

📊 Output data

A server found by keyword in the directory:

{
  "id": "143852930036924417",
  "url": "https://discord.com/servers/wynncraft-mmorpg-minecraft-server-143852930036924417",
  "name": "Wynncraft • MMORPG Minecraft Server",
  "description": "Full-Fat MMORPG Minecraft Server. No mods required! Join at (IP): play.wynncraft.com",
  "slug": "wynncraft-mmorpg-minecraft-server-143852930036924417",
  "category": "Gaming",
  "categories": ["Fandom", "Entertainment", "Memes", "Gaming", "General Chatting"],
  "keywords": ["grian", "minecraft", "minecraft mmorpg", "minecraft rpg", "minecraft server", "mmo", "mmorpg", "rpg", "winncraft", "wynn"],
  "features": ["COMMUNITY", "DISCOVERABLE", "ENABLED_DISCOVERABLE_BEFORE", "PARTNERED", "PREVIEW_ENABLED", "WELCOME_SCREEN_ENABLED"],
  "isVerified": false,
  "isPartnered": true,
  "isDiscoverable": true,
  "isCommunity": true,
  "isNsfw": false,
  "verificationLevel": null,
  "nsfwLevel": null,
  "memberCount": 218859,
  "onlineCount": 73515,
  "boostCount": 60,
  "boostTier": null,
  "locale": "en-US",
  "createdAt": "2016-02-01T23:00:11Z",
  "inviteUrl": "https://discord.gg/wynncraft",
  "vanityUrl": "https://discord.gg/wynncraft",
  "iconUrl": "https://cdn.discordapp.com/icons/143852930036924417/a_8f1e212da56e0f5b02ed58d65cdcf6cb.png?size=256",
  "bannerUrl": "https://cdn.discordapp.com/banners/143852930036924417/a_e9bb762b9c20ab5b83658671a70b3b28.gif?size=1024",
  "splashUrl": "https://cdn.discordapp.com/discovery-splashes/143852930036924417/9de202132b04423ac657c9716a11d8b1.jpg?size=2048",
  "socialLinks": null,
  "reasonsToJoin": null,
  "about": null,
  "foundBy": "keyword",
  "keyword": "minecraft",
  "position": 1,
  "totalResults": 4457,
  "scrapedAt": "2026-09-23T15:25:58Z"
}

A server resolved from an invite link:

{
  "id": "267624335836053506",
  "url": "https://discord.com/servers/python-267624335836053506",
  "name": "Python",
  "description": "We're a large community focused around the Python programming language. We believe that anyone can learn to code.",
  "slug": "python-267624335836053506",
  "category": null,
  "categories": [],
  "keywords": [],
  "features": ["ANIMATED_ICON", "BANNER", "COMMUNITY", "DISCOVERABLE", "NEWS", "VANITY_URL", "VERIFIED", "WELCOME_SCREEN_ENABLED"],
  "isVerified": true,
  "isPartnered": false,
  "isDiscoverable": true,
  "isCommunity": true,
  "isNsfw": false,
  "verificationLevel": 2,
  "nsfwLevel": 0,
  "memberCount": 431765,
  "onlineCount": 43054,
  "boostCount": 28,
  "boostTier": 3,
  "locale": null,
  "createdAt": "2017-01-08T12:03:33Z",
  "inviteUrl": "https://discord.gg/python",
  "vanityUrl": "https://discord.gg/python",
  "iconUrl": "https://cdn.discordapp.com/icons/267624335836053506/95a69567efa60a48b8d9840cb38b106b.png?size=256",
  "bannerUrl": "https://cdn.discordapp.com/banners/267624335836053506/ed3be5b67371fe219d7006152087afdb.png?size=1024",
  "splashUrl": "https://cdn.discordapp.com/splashes/267624335836053506/b7099fe221153a7faff76048dc2e63ef.jpg?size=2048",
  "socialLinks": null,
  "reasonsToJoin": null,
  "about": null,
  "foundBy": "invite",
  "keyword": null,
  "position": null,
  "totalResults": null,
  "scrapedAt": "2026-09-23T15:27:41Z"
}

The features list above is shortened; a real row lists every feature the server has enabled, often 40 or more.

🗂️ Data fields

35 fields per server. A directory row fills 28 of them on every row; an invite row adds verificationLevel, nsfwLevel and boostTier; a server page adds about, reasonsToJoin and socialLinks.

Group	Fields
Identity	id (server id, stable), url (directory page, or the invite link for a server outside the directory), name, description, slug
Content	category (primary), categories[], keywords[] (tags set by the server), features[] (Discord feature flags), isVerified, isPartnered, isDiscoverable, isCommunity, isNsfw, verificationLevel (0 to 4, invites only), nsfwLevel (0 to 3, invites only)
Measures	memberCount, onlineCount (members present when the page was read), boostCount (Nitro boosts), boostTier (0 to 3, invites only)
Place and time	locale (server language, en-US, pt-BR), createdAt (RFC 3339, from the server id)
Links and media	inviteUrl (vanity invite of a directory server, or the invite you gave), vanityUrl, iconUrl, bannerUrl, splashUrl, socialLinks[], reasonsToJoin[], about (the last three with includeDetails or a server page)
Meta	foundBy (keyword, category, directory, url, invite), keyword, position (rank in the listing, 1 = largest), totalResults (servers matching the search), scrapedAt

Usability facts:

• url is permanent for a directory server: any page address ending with the server id resolves, and the row carries the canonical one.
• iconUrl, bannerUrl and splashUrl stay valid as long as the server keeps that image; an animated icon or banner is a .gif.
• inviteUrl of a directory server is its vanity link, which Discord grants to every discoverable server. An invite you paste may expire; a dead one gives no row and is counted in the run summary.
• memberCount and onlineCount are the approximate counts Discord shows; onlineCount is a snapshot of the moment the page was read.
• The keywords field is the server's own tags. Your search keyword is in keyword.

🚀 Examples

Get the largest Minecraft servers

{
  "keywords": ["minecraft"],
  "maxItems": 100
}

List the top 500 gaming servers

{
  "category": "gaming",
  "maxItems": 500
}

Search anime servers inside Entertainment with their full pages

{
  "keywords": ["anime"],
  "category": "entertainment",
  "includeDetails": true,
  "maxItems": 50
}

Resolve a list of invite links

{
  "startUrls": [
    "https://discord.gg/python",
    "https://discord.com/invite/discord-developers",
    "midjourney"
  ],
  "maxItems": 50
}

Read a directory search copied from the browser

{
  "startUrls": ["https://discord.com/servers/music?query=lofi"],
  "maxItems": 30
}

Cover a market with several keywords in one run

{
  "keywords": ["crypto", "trading", "nft"],
  "maxItems": 300
}

Map the whole directory, largest servers first

{
  "maxItems": 1000
}

🤖 Copy to your AI assistant

Paste this block into Claude, ChatGPT or Cursor to give it full context about this scraper:

You have access to the Discord Scraper on Apify: silentflow/discord-scraper

Input schema:
- keywords (array of strings, optional): words searched in Discord's server directory, largest servers first, up to 3,000 servers per keyword
- category (string, optional): "gaming", "music", "entertainment", "education" or "" for all; narrows keywords or lists the category alone
- startUrls (array of strings, optional): server pages (https://discord.com/servers/<slug>-<id>), directory search pages, invite links (https://discord.gg/<code>, discord.com/invite/<code>) or bare invite codes
- maxItems (integer, default 100): total servers to deliver
- includeDetails (boolean, default false): read each server's page for about, reasonsToJoin, socialLinks and the full features list
- debugMode (boolean, default false)
With no keyword, category or URL the run lists the whole directory by member count.

Output per server (35 fields):
- id (string), url (string), name (string), description (string|null), slug (string|null)
- category (string|null), categories (string[]), keywords (string[]), features (string[])
- isVerified, isPartnered, isDiscoverable, isCommunity, isNsfw (boolean)
- verificationLevel (int|null), nsfwLevel (int|null): invites only
- memberCount (int|null), onlineCount (int|null), boostCount (int|null), boostTier (int|null, invites only)
- locale (string|null), createdAt (RFC 3339 string|null)
- inviteUrl, vanityUrl, iconUrl, bannerUrl, splashUrl (string|null)
- socialLinks (string[]|null), reasonsToJoin (string[]|null), about (string|null): server page only
- foundBy ("keyword"|"category"|"directory"|"url"|"invite"), keyword (string|null), position (int|null), totalResults (int|null), scrapedAt (RFC 3339 string)

One event per server row. No Discord account needed. Use apify-client for Python or JS.

💻 Integrations

Build a partnership outreach list in Python

from apify_client import ApifyClient
import csv

client = ApifyClient("YOUR_APIFY_TOKEN")
run = client.actor("silentflow/discord-scraper").call(run_input={
    "keywords": ["anime", "manga"],
    "category": "entertainment",
    "maxItems": 300,
})
rows = client.dataset(run["defaultDatasetId"]).list_items().items

# Active communities only: at least 2,000 members online, ranked by online count
active = sorted(
    (r for r in rows if (r["onlineCount"] or 0) >= 2000),
    key=lambda r: r["onlineCount"], reverse=True,
)
with open("anime-outreach.csv", "w", newline="") as f:
    w = csv.writer(f)
    w.writerow(["name", "members", "online", "boosts", "created", "invite", "page"])
    for r in active:
        w.writerow([r["name"], r["memberCount"], r["onlineCount"], r["boostCount"], r["createdAt"][:10], r["inviteUrl"], r["url"]])
print(f"{len(active)} active servers out of {len(rows)}")

Track the growth of your market's servers every week in Node.js

import { ApifyClient } from "apify-client";
import fs from "node:fs";

const client = new ApifyClient({ token: "YOUR_APIFY_TOKEN" });
const run = await client.actor("silentflow/discord-scraper").call({
    startUrls: ["https://discord.gg/python", "https://discord.gg/discord-developers", "midjourney"],
    maxItems: 50,
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();

const previous = fs.existsSync("servers.json") ? JSON.parse(fs.readFileSync("servers.json")) : {};
for (const s of items) {
    const before = previous[s.id]?.memberCount;
    const delta = before ? s.memberCount - before : 0;
    console.log(`${s.name}: ${s.memberCount} members (${delta >= 0 ? "+" : ""}${delta} since last week), ${s.onlineCount} online`);
    previous[s.id] = { memberCount: s.memberCount, onlineCount: s.onlineCount, at: s.scrapedAt };
}
fs.writeFileSync("servers.json", JSON.stringify(previous, null, 2));

Export a category to CSV from the command line

apify call silentflow/discord-scraper --input '{"category": "gaming", "maxItems": 500}' --wait-for-finish
apify datasets get-items --format csv --output gaming-servers.csv

📈 Performance

Run	Result	Time measured
Default input, keyword minecraft, 100 servers	100 rows, 28 fields filled on every row	14 seconds
Category music, 30 servers	30 rows	5 seconds
Keyword anime in Entertainment with includeDetails, 15 servers	15 rows with about, social links and full features	28 seconds
Two invites, one server page and a directory search, 30 servers	30 rows	9 seconds
A whole category, 500 servers	500 rows	about 90 seconds
The whole directory, 1,000 servers	1,000 rows	about 3 minutes

Memory: 256 MB is enough for any run. The directory serves 12 servers per page and stops at 3,000 servers per search; the run reads one page every second or two, which is the rhythm the directory accepts from one connection.

💾 Data export

Every run's dataset is available on the Storage tab of the run as JSON, CSV, Excel, XML, HTML table or RSS, and through the API:

https://api.apify.com/v2/datasets/{DATASET_ID}/items?format=csv&token=YOUR_TOKEN

Replace csv with json, xlsx or xml. Add &fields=name,memberCount,onlineCount,inviteUrl to keep only the columns you need, and &clean=true to drop empty values.

💡 Tips for best results

1. Start with a category when the keyword is broad. minecraft alone matches 4,457 servers and gaming 38,454; the directory stops at 3,000 per search, so keywords: ["rpg"] with category: "gaming" reaches deeper than rpg alone.
2. Use several precise keywords instead of one vague one. ["genshin", "honkai", "zenless"] in one run returns each franchise's servers, deduplicated across the three searches.
3. Rank on onlineCount, not only memberCount. A server with 500,000 members and 3,000 online is quieter than one with 60,000 members and 8,000 online. position gives the directory's own ranking by members.
4. Leave includeDetails off for discovery, on for the shortlist. The directory row already carries counts, tags, language and the invite link; the page adds the long description, reasons to join and social links at about two seconds per server. Discover with 1,000 rows, then run the 50 servers you keep with details.
5. Schedule the same input weekly and compare memberCount by id. The id never changes, the name sometimes does. createdAt tells you whether a fast-growing server is new or established.

❓ FAQ

What does this scraper extract? Public Discord servers, one row each: name, description, member and online counts, boosts, category and categories, tags, enabled features, language, creation date, invite link, icon, banner and splash, plus the reasons to join, social links and long description when a server page is read.

How do I find servers? Four ways: a keyword searched in Discord's directory, a category browsed by size, a server page or a directory search URL pasted from the browser, and an invite link or code. They combine in one run and the rows are deduplicated by server id.

Do I need a Discord account? No. The scraper reads the public server directory and the public invite preview. No login, no token, nothing of yours is used.

How fresh are the numbers? Live: every run reads the directory pages and invite previews at run time, nothing is cached. memberCount and onlineCount are the approximate counts Discord shows at that moment.

Which servers can it find? Every server Discord lists in its directory, 52,816 on 23 September 2026, by keyword or category. Any other public server is reachable through its invite link, including servers that never applied to the directory.

Is there a limit per search? The directory answers at most 3,000 servers for one keyword or category, 12 per page. Narrow a broad keyword with a category or split it into precise ones. maxItems caps the run and has no fixed limit.

Why is category empty on some rows? A server resolved from an invite carries no category: the invite preview does not include it. Directory rows always have one. The same goes for keyword, position and totalResults, which describe a search.

Which categories exist? The directory has four browseable categories: Gaming, Music, Entertainment and Education. Other categories such as Science & Tech, Creative Arts or Finance appear in categories on every row but have no page of their own; search them with a keyword.

Can I get member lists or messages? No. This scraper delivers servers. Member lists and message history need a logged-in Discord account and are out of scope.

Do keyword and category combine? Yes. keywords: ["rpg"] with category: "gaming" returns the RPG servers of the Gaming category only, largest first, and reaches past the 3,000-server limit of a plain rpg search.

Do the links expire? url, iconUrl, bannerUrl and splashUrl stay valid as long as the server keeps its page and images. The vanity inviteUrl of a directory server is permanent; an invite you paste may expire, in which case the run reports it and delivers no row for it.

Can I run several keywords at once? Yes. Every keyword is searched in turn, the rows are deduplicated across searches, and maxItems applies to the whole run.

What is discord.com/servers? Discord's own server directory, where community servers apply to be discoverable. It is public, ranked by member count, and searchable by keyword and category. This scraper reads it directly.

⚖️ Legal

This Actor extracts publicly available data from Discord's server directory and public invite previews. It does not bypass any login, paywall or CAPTCHA, and never uses a user account. Users are responsible for complying with Discord's terms of service and applicable data protection laws (GDPR, CCPA, and PIPL where relevant). The output describes servers, not people; server names and descriptions may mention individuals, and users must handle such data accordingly. The data returned is informational; verify accuracy for regulated use cases.

🔗 Related scrapers

• YouTube Channel Scraper: channels with subscribers, views and recent videos, for the same creators' other communities.
• TikTok Scraper: videos, profiles and hashtags with engagement counts.
• Instagram Engagement Scraper: likes and comments of any public post.
• Facebook Search Scraper: pages, groups and posts found by keyword.

📬 Support

Need something this scraper doesn't do yet? We ship features fast.

• Feature requests go straight to our backlog
• Enterprise needs? We do custom integrations and high-volume plans
• Pricing questions? Check the Monetization tab on the actor page

Response time: usually under 24 hours.

Check out our other scrapers: silentflow on Apify