Sherlock Username Scraper

Search for a username across 400+ social networks and websites. Input one or more usernames and get back every platform where that username exists, with direct profile URLs. Built for OSINT research, due-diligence, lead enrichment, and social media audits.

---

What this actor does

For each username you provide, the actor:

1. Loads the Sherlock OSINT site database — 480+ social network definitions from the open-source Sherlock project, fetched fresh on each run
2. Checks each platform concurrently — makes an HTTP request to the expected profile URL on every site, using per-site detection rules
3. Applies smart detection logic — each site uses one of three methods to confirm presence: HTTP status code check, response body text check, or redirect URL check
4. Returns only confirmed hits — saves a record for each platform where the username is found, with the direct profile URL

Typical run: one username against 480 platforms takes under 60 seconds. Profiles found on 50–200 platforms is normal for common usernames.

---

Input

Field	Type	Default	Description
usernames	array of strings	—	Required. Usernames to search. Each is checked against all platforms.
maxItems	integer	0	Maximum number of found profiles to return across all usernames. 0 = no limit.
concurrency	integer	20	Concurrent HTTP requests per username. Range 10–50 recommended.
timeout	integer	10000	HTTP request timeout per site check in milliseconds.

Example input:

{
  "usernames": ["johnsmith", "john_smith_dev"],
  "maxItems": 0,
  "concurrency": 20,
  "timeout": 10000
}

---

Output

One record per platform where the username was found.

Field	Type	Description
username	string	The username that was searched
site_name	string	Name of the social network or website
url	string	Direct URL to the profile
url_main	string	Main URL of the social network
status	string	found — only found profiles are saved
http_status	number	HTTP status code returned by the site
response_time_ms	number	Response time in milliseconds
scraped_at	string	ISO timestamp when the check was performed

Example output record:

{
  "username": "johnsmith",
  "site_name": "GitHub",
  "url": "https://github.com/johnsmith",
  "url_main": "https://github.com/",
  "status": "found",
  "http_status": 200,
  "response_time_ms": 312,
  "scraped_at": "2026-06-02T10:15:42.000Z"
}

---

Notes

• Platform coverage: 480+ sites including GitHub, Twitter/X, Instagram, Reddit, LinkedIn, YouTube, TikTok, Pinterest, Twitch, Medium, and hundreds more.
• Detection accuracy: Uses the same detection rules as the open-source Sherlock tool. Some sites may produce false positives (username check returns 200 for non-existent accounts). Cross-reference with the url field to verify.
• No authentication required: All checks are against public-facing profile pages.
• No proxy needed: Requests are spread across hundreds of different domains at low concurrency — no single site sees more than one request per run.
• Concurrency: Increase concurrency (up to 50) for faster runs; decrease if you see timeouts on slow platforms. Default of 20 balances speed and reliability.