arxhr007/Aliens_eye

Hunt down 840+ social media accounts using AI

What it solves

Aliens Eye is an AI-powered OSINT (Open Source Intelligence) tool designed to find social media profiles across hundreds of platforms using a single username. It solves the problem of "false positives" common in basic username scanners—where a site might return a success code even if a profile doesn't exist—by using machine learning to distinguish real profiles from error pages or bot challenges.

How it works

The tool scans over 840 platforms asynchronously. Instead of relying solely on HTTP status codes, it converts every page response into a 30-dimensional feature vector (analyzing DOM structure, keywords, and metadata). A blended detection system then votes on the result:

  1. Heuristic Engine: Uses weighted scoring based on the feature vector.
  2. ML Model: A logistic regression model trained on real and fake accounts to provide a probability score.

To further ensure accuracy, it performs a "control check" by requesting a username that cannot exist; if the site returns the same page, the original hit is downgraded. It also supports profile extraction (bios, avatars), cross-site correlation to cluster profiles belonging to the same person, and recursive scanning of usernames found in bios.

Who it’s for

It is built for OSINT researchers, investigators, and security professionals who need to accurately map a target's digital footprint across the web.

Highlights

  • ML-Blended Detection: Combines machine learning with 30 structural signals to reduce false positives.
  • Massive Coverage: Scans 840+ platforms in seconds.
  • Advanced Correlation: Clusters profiles by avatar hashes, shared links, and bios.
  • Recursive Expansion: Automatically follows and scans usernames discovered within profile bios.
  • LLM Integration: Includes an MCP server to expose scanning capabilities to AI agents.
  • Comprehensive Reporting: Exports results to JSON, CSV, HTML, PDF, and graph formats (e.g., Maltego, Mermaid).
  • Retrainable Model: Allows users to collect new data and retrain the ML model via a built-in training pipeline.

Related

  • Project
    kaifcodec/user-scanner

    An OSINT suite for deep email and username intelligence that maps digital footprints across 2,700+ platforms with automated pivoting and AI agent integration.

  • Project
    0x6rss/instagram-private-graph

    A local OSINT research tool for Instagram that analyzes account relationships using a repeated-appearance model to generate ranked candidate lists and network visualizations.

  • Project
    OpenOSINT/OpenOSINT

    OpenOSINT is an MIT‑licensed Python package that lets security analysts run 21 OSINT utilities (email, username, breach, WHOIS, IP, DNS, etc.) via a natural‑language LLM interface. It can be used as a REPL, CLI, web chat UI, or as an MCP server for programmatic access. By default it operates in a passive mode (only read‑only API calls); active, potentially noisy tools must be explicitly enabled. The toolset is extensible, supports parallel execution, auto‑generates Markdown/PDF reports, and offers a pay‑per‑use cloud version on Apify. Install with `pip install openosint` and run `openosint` (or `openosint web`).

  • Project
    SleepySoft/IntelligenceIntegrationSystem

    An OSINT system that scrapes public news from global media outlets and uses AI to analyze, score, and filter out low-value content to extract high-value intelligence.

  • Project
    KatrielMoses/voidaccess

    A self-hostable OSINT tool that automates dark-web research by collecting, enriching, and mapping threat intelligence into structured formats.