Cybersecurity / Digital Forensics 2026
Forensic File Analyzer
A learning project in digital forensics: upload a file and read its hashes, signature, entropy, strings and heuristic indicators.
- Year
- 2026
- Status
- Learning project
- Platform
- Web · API
- Scope
- Web application, Analysis API
- 01UploadStreamed to temporary storage
- 02HashesMD5, SHA-1, SHA-256, SHA-512
- 03SignatureMagic number versus extension
- 04EntropyOverall and per chunk
- 05StringsURLs, IPs, emails, paths
- 06IndicatorsRule-based score, 0–100
Overview
A small full-stack application for practising digital forensics concepts. A file is streamed to temporary storage, run through an analysis pipeline, deleted, and the results are returned as a dashboard. It is an educational tool, not a production security product.
Challenge
Forensic concepts — magic numbers, entropy, string extraction — are easy to read about and hard to build intuition for. Seeing them computed on real files makes the difference.
Direction
Keep each analysis a small, separate service so it can be read and tested on its own, and frame every heuristic honestly: the risk score is a transparent rule-based indicator, never a malware verdict.
What was built
- MD5, SHA-1, SHA-256 and SHA-512, computed in chunks so large files never load into memory
- Signature detection against a table of common formats, with extension-mismatch flags
- Overall and chunk-by-chunk Shannon entropy, charted
- String extraction with detection of URLs, IP addresses, emails and paths
- A rule-based heuristic score from 0 to 100, with each rule shown
- Multi-file scan with duplicate grouping by SHA-256
- Report export as JSON, CSV or a self-contained HTML file
Technology
- React
- Vite
- Tailwind CSS
- FastAPI
- Python
- pytest
Outcome
A working analysis pipeline and dashboard for studying file forensics, with uploads deleted as soon as each request completes.
Technical notes
Scope
Built for education and forensic research practice. Its heuristic indicators are not malware detection and not forensic conclusions.