Poeth01/SecureAI-PolicyGuard
🛡️ Protect enterprise data with SecureAI PolicyGuard, an AI-driven platform for automated classification and compliance management.
What's novel
🛡️ Protect enterprise data with SecureAI PolicyGuard, an AI-driven platform for automated classification and compliance management.
Code Analysis
12 files read · 3 roundsA Tkinter desktop app that scans .txt files in a directory, uses regex to detect PII (emails, IBANs, credit cards), applies a keyword-based risk score to classify files into 4 tiers, optionally encrypts them with Fernet, and logs actions to SQLite, with an optional FastAPI server exposing the same l
Strengths
Clean separation of concerns across small, focused modules (classifier, policy_engine, crypto_engine, storage, ner_engine, risk_engine) with readable code, proper error handling with fallbacks, and a well-structured FastAPI layer with Pydantic models. The localization system is thorough and the UI is functional.
Weaknesses
The README is deeply misleading, claiming 'AI-powered enterprise data protection' when the code is regex + keyword matching with no ML whatsoever, and links to a zip file as a 'download' in a pattern consistent with malware distribution. The Fernet encryption key is stored in plaintext in the same directory as the data it protects, the API has no authentication, and there are zero tests despite test frameworks in requirements.txt.
Score Breakdown
Signal breakdown
Innovation
Craft
Traction
Scope
Evidence
Commits
54
Contributors
2
Files
19
Active weeks
3
Repository
Language
Python
Stars
1
Forks
0
License
MIT