SHERLOCK-RS

Hunt social media accounts by username — Rust Edition

A complete Rust rewrite of Sherlock with a modern dark web UI — single .exe, zero installation.

---

Overview

Sherlock-RS scans 478+ social platforms in parallel to check whether a username exists. Just run the exe: a local server starts, your browser opens automatically, and results stream in real time.

New: Simultaneous multi-username search with tabs, automatic rotation of 25 real User-Agents, and smart retry logic on network errors.

---

Features

Feature	Details
🔍 478+ sites scanned	Official Sherlock database, one-click update from the UI
👥 Multi-username	Enter multiple names at once (comma or newline), results per tab
⚡ Parallel scanning	20 concurrent requests via Tokio async — full scan in minutes
🔄 User-Agent rotation	25 real browsers (Chrome, Firefox, Edge, Safari, Opera…) rotated randomly per request
🔁 Smart retry	3 attempts with exponential backoff (500 ms / 1 000 ms) on network errors only
🎨 Modern UI	Dark-themed web interface with real-time results (Server-Sent Events)
🛡️ WAF detection	Cloudflare, PerimeterX, AWS CloudFront detected and flagged
🧅 Proxy / Tor	Native SOCKS5 support (socks5://127.0.0.1:9050 for Tor)
📥 Export	Download results as CSV (spreadsheet) or TXT
🔎 Filter & sort	Sort by name, status or response time — live text filter
📦 Zero install	Single self-contained 5 MB .exe, no dependencies required

---

Installation

Quick method — Download the binary

1. Download the latest release from the Releases page
2. Double-click sherlock-rs.exe
3. Your browser opens automatically — that's it

Build from source

Requirements:

• Rust (install via rustup)
• Visual Studio Build Tools 2022 with the Desktop development with C++ workload

git clone https://github.com/Oli97430/sherlock-rs.git
cd sherlock-rs
cargo build --release

The binary will be located at target/release/sherlock-rs.exe.

---

Usage

sherlock-rs.exe

The program starts a local server on a random port and opens your default browser. No extra commands needed.

Basic steps

1. Enter one or more usernames to search (separated by comma or newline)
2. Adjust options if needed (timeout, proxy, NSFW)
3. Click Hunt or press Enter
4. Results appear in real time, site by site
5. Export results as CSV or TXT via the dedicated buttons

Multi-username search

The input field accepts multiple names at once:

johndoe
janedoe, alice

Each username gets its own tab with a live counter of accounts found. The tab being scanned pulses in blue.

Interface options

Option	Description
Timeout	Maximum wait time per request (default: 30 s, min: 5 s)
NSFW	Include adult content platforms in the search
Proxy	SOCKS5 or HTTP proxy URL, e.g. socks5://127.0.0.1:9050 for Tor
Update DB	Downloads the latest site database from GitHub

Keyboard shortcuts

Key	Action
Enter	Start search (from the username field)
Shift + Enter	Add a new line (multi-username input)
Escape	Stop the current search

---

How it works

Detection methods

Sherlock-RS faithfully implements the 3 detection methods from the original project:

Type	Logic
status_code	HTTP 404 (or custom code) → not found; 200-299 → found
message	Specific error text found in response body → not found
response_url	Redirects disabled; 200-299 → found, otherwise not found

WAF detection (Cloudflare, PerimeterX…) is applied first, before any other logic, to avoid false positives. Blocked results are reported separately with the Blocked status.

User-Agent rotation

Each individual request randomly picks a User-Agent from 25 real modern browsers:

• Chrome 128–131 (Windows, macOS, Linux, Android)
• Firefox 130–133 (Windows, macOS, Linux)
• Edge 130–131 (Windows, macOS)
• Safari 17 (macOS, iOS)
• Opera 116, Brave

This significantly reduces bot-detection-based blocking.

Exponential backoff retry

On network errors (timeout, connection refused, DNS failure):

Attempt 1  →  fails  →  wait 500 ms
Attempt 2  →  fails  →  wait 1 000 ms
Attempt 3  →  final result (success or error displayed)

Valid HTTP responses (even 403 or 404) do not trigger a retry.

---

Result statuses

Status	Meaning	Tip
✅ Found	Account detected on the platform	Click the URL to open the profile
❌ Not found	No account under this name	—
⚠️ Blocked	Blocked by a WAF (Cloudflare…)	Retry with a proxy or Tor
🔴 Error	Network error or timeout after 3 attempts	Increase the timeout
⬜ Invalid	Username format doesn't match the site's rules	Normal for some sites

---

Code architecture

sherlock-rs/
├── Cargo.toml              # Dependencies and project metadata
├── src/
│   ├── main.rs             # Entry point, console banner, server startup
│   ├── server.rs           # Axum server: REST routes + SSE streaming
│   ├── checker.rs          # Async scan engine: UA rotation, retry, detection
│   ├── sites.rs            # Load and parse data.json (local cache + GitHub)
│   ├── result.rs           # Types: QueryStatus (enum), QueryResult (struct)
│   └── export.rs           # CSV and TXT export grouped by username
└── frontend/
    └── index.html          # Full UI embedded in the binary (HTML/CSS/JS)

Rust crates used

Role	Crate
Async runtime	tokio 1
Web server + SSE	axum 0.7
HTTP client	reqwest 0.12
JSON serialization	serde + serde_json
Regular expressions	regex
Randomness (UA rotation)	rand 0.8
CSV export	csv
Browser launcher	open
System directories	dirs
Error handling	anyhow

---

Credits

• Author: Olivier Hoarau — tarraw974@gmail.com
• Original project: Sherlock Project by @sdushantha and the community (MIT license)
• Database: data.json maintained by the Sherlock Project community

---

License

MIT — see the LICENSE file

---

_{Built with passion and Rust 🦀 — Olivier Hoarau}