Multi-Site Username Lookup & Profile Link Finder

Multi-Site Username Lookup & Profile Finder

Find where a username exists across 400+ social networks, developer platforms, forums, gaming sites and marketplaces, and get a real profile URL instead of a guess. Every result carries a per-site status, HTTP code, detection method and verification flag, so a soft-404 page can never masquerade as a profile.

What this actor does

• Checks up to 100 usernames per run with concurrent site checks.
• Uses 22 native probes with exact API/XML/HTML detection for GitHub, GitLab, Steam, Chess.com, Lichess, Hacker News, Codeforces, Docker Hub, Roblox, YouTube, Mastodon, Dev.to, SoundCloud, Instagram, Pinterest, Twitch, X, LeetCode, Keybase, Stack Exchange, Dailymotion and Kick.
• Runs decoy verification on the generic long tail: a site that claims a random handle is marked unverified and kept out of links.
• Exposes site, category, status, HTTP status, detection method and response time for every check, plus an NSFW flag on adult-site rows.
• Extracts profile fields where the site exposes them: display name, bio, avatar, followers, following, posts, join date, location, website and verified badge.
• Hides NSFW sites by default and removes ten sites proved unable to distinguish real profiles from missing ones.
• Supports one row per username (legacy-compatible), one row per site, or one row per claimed profile.
• Needs no login, cookie, API key or proxy.

Input

Output

perUsername (default, backward compatible)

One row per username. username and links[] keep their original meaning; links now contains only verified claimed profiles. New summary fields and a profiles[] array sit beside them.

{
  "username": "octocat",
  "links": ["https://github.com/octocat"],
  "sitesChecked": 428,
  "profilesFound": 1,
  "sitesAvailable": 380,
  "sitesBlocked": 12,
  "sitesUnknown": 15,
  "sitesUnverified": 0,
  "checkedAt": "2026-11-04T12:00:00Z",
  "profiles": [
    {
      "username": "octocat",
      "site": "GitHub",
      "category": "dev",
      "url": "https://github.com/octocat",
      "status": "claimed",
      "httpStatus": 200,
      "detectionMethod": "api_json",
      "verified": true,
      "responseTimeMs": 210,
      "checkedAt": "2026-11-04T12:00:00Z",
      "profile": {
        "userId": "583231",
        "username": "octocat",
        "displayName": "The Octocat",
        "avatarUrl": "https://avatars.githubusercontent.com/u/583231",
        "followers": 18432,
        "following": 9,
        "joinedAt": "2011-01-25T18:44:36Z",
        "location": "San Francisco",
        "website": "https://github.blog",
        "stats": { "repositories": 8, "gists": 8 }
      }
    }
  ]
}

perSite

One row per username × site with username, site, category, url, status, httpStatus, detectionMethod, responseTimeMs and checkedAt, plus verified on hit rows, isNsfw on adult rows and profile when enrichment is on. By default only claimed and unverified rows are emitted; includeAvailable adds the rest.

perHit

One row per claimed and verified profile, with the same shape as perSite. With verifyClaims=false, claimed rows are emitted without the verification gate.

Status vocabulary

Empty strings, arrays and objects are omitted at every depth.

Examples

Check a batch of handles across every non-NSFW site:

{
  "usernames": ["johndoe", "janedoe", "octocat"],
  "maxUsernames": 3
}

Check only developer and gaming sites, one row per hit:

{
  "usernames": ["octocat"],
  "categories": ["dev", "gaming"],
  "outputMode": "perHit",
  "enrichProfiles": true
}

Availability check for a new handle on selected sites:

{
  "usernames": ["mynewbrand2026"],
  "sites": ["GitHub", "Instagram", "X (Twitter)", "Twitch", "Reddit"],
  "includeAvailable": true,
  "outputMode": "perSite",
  "verifyClaims": true
}

Fast top-20 scan per handle:

{
  "usernames": ["johndoe"],
  "maxSitesPerUsername": 20,
  "enrichProfiles": false
}

Use cases

• Brand protection — find impersonating or parked accounts across networks.
• OSINT and investigations — build a verified footprint from one handle.
• Lead enrichment — attach public profile URLs and display names to a handle list.
• Username availability — see where a new brand handle is free before registering.
• Recruiting and sourcing — locate developer profiles on GitHub, GitLab, Codeforces, LeetCode and Stack Exchange.
• Creator discovery — compare presence across YouTube, Twitch, Instagram, Pinterest and X.

FAQ

How do I know a hit is real? Native probes use exact API/XML/HTML signals. Generic sites are re-checked with a random decoy handle: if the site claims the decoy too, the original row becomes unverified and is excluded from links. The detectionMethod and verified fields tell you exactly which path proved each hit.

Why are there fewer links than the old version? The old actor reported soft-404 and WAF pages as profiles. Ten such sites were removed from the catalog, and the remaining generic claims now pass the decoy check. Set verifyClaims=false to see the raw claimed rows.

What does unverified mean? The site returned a claimed signal, but a random handle produced the same signal, so the site cannot distinguish real profiles. The row is kept for transparency (and in perSite output) but never enters links. profilesFound counts claimed plus unverified candidates; sitesUnverified and links show which are proven.

Why are NSFW sites missing? They are excluded unless includeNsfw=true. The catalog marks 16 adult entries.

A site I need is missing. Can I add it? The catalog is pinned and covers 428 sites. Use sites to confirm whether a site is supported; unknown values are logged and ignored. The catalog can be refreshed with scripts/refresh_catalog.py when upstream adds sites.

How does {?} work? It expands to three handles using _, - and .. For example john{?}doe checks john_doe, john-doe and john.doe. Expansion happens before maxUsernames is applied.

Do I need cookies, an API key or a proxy? No. The actor is cookieless and anonymous. Sites that hard-block datacenter IPs or serve a bot challenge are reported honestly as waf or unknown rather than guessed.

Why do some sites return unknown or waf? Cloudflare, Akamai and similar edges sometimes challenge anonymous requests. Retries handle transient 429/5xx responses, but a persistent challenge stays waf, and a timeout or 5xx stays unknown.

Can I check 100 handles in one run? Yes, maxUsernames accepts 1–100. Handles are processed four at a time, with up to 50 site checks in flight across the run. Large scans take longer; narrow them with sites, categories or maxSitesPerUsername. GitHub's anonymous API allows 60 requests/hour, so the tail of a 100-handle GitHub-only batch can fall back to HTML or report waf.

Which fields can be enriched? Display name, bio, avatar, followers, following, posts, join date, location, website and verified badge, where the site's public response exposes them. Long-tail sites return URL and status only.

Limitations

• No login-only data. Private profiles, follower lists, posts and activity feeds are out of scope.
• Some sites cannot be detected anonymously. Reddit, TikTok and Spotify are not shipped as claimed paths; they are either excluded upstream or return waf/unknown. Sites behind a persistent bot challenge behave the same way.
• Instagram and X throttle HTML. From datacenter IPs they can serve a login or generic page, so an existing profile may be reported available or unknown; no hit is fabricated.
• Native enrichment covers 22 sites. The long tail provides URL, status, HTTP code and detection method only.
• GitHub's anonymous API is limited to 60 requests/hour. Large GitHub-only batches fall back to HTML (fewer profile fields) or report waf for the tail.
• Catalog size is 428 sites. This is a footprint tool, not a 3,000-site Maigret replacement.
• Decoy verification trades recall for precision. A site that claims every handle is reported as unverified, even if a real profile exists there.