Spaces:
Runtime error
Runtime error
Download DETAILS.MD from sRjN17/Entity-Discovery: direct link, hf CLI and curl.
- Browser
- Download file 2.86 kB
-
https://huggingface.co/spaces/sRjN17/Entity-Discovery/resolve/main/DETAILS.MD
- Command line
-
hf download hf://spaces/sRjN17/Entity-Discovery/DETAILS.MD
-
curl -L -o DETAILS.MD https://huggingface.co/spaces/sRjN17/Entity-Discovery/resolve/main/DETAILS.MD
2.86 kB
| Technical Assessment | |
| **1\. Project Overview** | |
| The objective is to develop a prototype OSINT (Open-Source Intelligence) tool capable of automated entity discovery and reporting using Python. The tool must accept a Company Name or Individual Name as input, aggregate data from disparate digital sources, and synthesize the findings into a professional investigation report. | |
| **2\. Core Requirements** | |
| **Phase I: Data Acquisition (Multi-Vector Search)** | |
| The system must implement a modular 'Adapter' pattern to pull data from at least three distinct source categories: | |
| - Social & Public Footprint: e.g., Automated Google Dorking, LinkedIn scraping, or Twitter/X mentions. | |
| - Technical Infrastructure: e.g., WHOIS records, DNS history, GitHub repositories, or breach status (HIBP). | |
| - Contextual & Regulatory: e.g., OpenCorporates API, news archives, or official government registries. | |
| **Phase II: Analysis & Disambiguation** | |
| Data without context is noise. The developer must implement logic to handle Entity Resolution: | |
| - Grouping: Link identified domains, IP addresses, and key employees to the parent company. | |
| - Validation: Differentiate between "False Positives" using contextual clues (location, industry). | |
| - Risk Scoring (Bonus): Implement a weighting algorithm to assign a Risk Score (e.g., Leaks = Critical). | |
| **Phase III: Reporting Engine** | |
| The system must generate a professional PDF or Markdown file containing: | |
| - Executive Summary: A narrative, high-level overview of the target's digital footprint. | |
| - Categorized Data Tables: Structured lists of discovered assets. | |
| - Audit Trail: Every data point must include a source URL and retrieval timestamp. | |
| **3\. Evaluation Criteria** | |
| | Criteria | Exceptional | Satisfactory | | |
| | -------------- | ------------------------------------------------- | --------------------------------------- | | |
| | Data Integrity | Handles "False Positives" with confidence scores. | Provides raw data with no filtering. | | |
| | --- | --- | --- | | |
| | Architecture | Highly modular (easy to add new sources). | Monolithic script logic. | | |
| | --- | --- | --- | | |
| | OPSEC | Respects robots.txt and uses proxy rotation. | Basic requests that get flagged easily. | | |
| | --- | --- | --- | | |
| **4\. Submission Deliverables** | |
| - Source Code: Hosted on a private GitHub/GitLab repository. | |
| - README: Documentation on setup, API keys, and architecture decisions. | |
| - Sample Report: A PDF generated by the tool for entity (e.g., "Travis Haasch" CEO of AIGeeks). |