-
Notifications
You must be signed in to change notification settings - Fork 0
Home
PATAS Bot edited this page Nov 19, 2025
·
21 revisions
Pattern-Adaptive Transmodal Anti-Spam System
PATAS Core technical documentation. This wiki contains technical documentation for integration and deployment.
- Quick Start Guide - Installation and basic usage
- Architecture - System architecture and design principles
- Code Overview - Code structure and navigation guide
- Configuration - Configuration reference and environment variables
- Scaling & Cost Design - How PATAS handles millions of logs efficiently
- Comparison and Positioning - PATAS vs traditional approaches
- Roadmap - Development roadmap and future enhancements
- Safety Profiles - Conservative, Balanced, and Aggressive profiles
- LLM Usage - LLM integration and privacy guarantees
- PATAS LLM Roadmap - Long-term LLM strategy and potential in-house model
- PATAS LLM Preparation Notes - Phase 0 implementation details
- Engineering Notes for Telegram - Technical overview for Telegram engineers
- Telegram Safety Guide - Safety profiles and enforcement model
- API Reference - Complete API endpoint documentation
- API Quickstart - Quick start guide for API usage
- Privacy and Data Protection - Privacy modes and data handling
- Security - Security considerations and best practices
- Security Audit Checklist - Comprehensive security checklist
- Deployment - Production deployment guide
- Load Testing Guide - Load testing tools and procedures
PATAS is an autonomous pattern discovery and rule management system for anti-spam operations. It analyzes historical message logs, automatically discovers spam patterns, generates safe blocking rules, and evaluates their effectiveness before deployment.
Key Characteristics:
- Signal engine, not enforcement - PATAS provides patterns and metrics that inform anti-spam decisions
- On-premise deployment - Designed for deployment within your infrastructure
- Two-stage processing - Fast scanning + deep analysis for 70-90% cost reduction
- Deterministic and rule-based - Core engine is deterministic; ML/LLM is optional
- Safety-first design - Multiple safety profiles with clear risk boundaries
- Ingest - Load historical message logs into PATAS
- Discover - Automatically identify recurring spam patterns (two-stage: fast scan + deep analysis)
- Generate - Create safe SQL rules from discovered patterns
- Evaluate - Test rules on historical data (shadow mode)
- Promote - Activate rules that meet safety thresholds
- Monitor - Track rule performance and deprecate underperforming rules
Architecture and Design:
- Start with Architecture for high-level understanding
- Review Code Overview to understand codebase structure
- See Configuration for setup and tuning
Scaling and Performance:
- Scaling & Cost Design - How PATAS scales efficiently
Telegram Integration:
- Start with Engineering Notes for Telegram
- Review Telegram Safety Guide
API and Operations:
- Review API Reference for integration
- See Configuration and Quick Start for deployment
- See Deployment for production setup
- GitHub: kiku-jw/PATAS