Enforces a structured, self-improving context and workflow system for AI agents using the Atris methodology.
Atris is a comprehensive context management and workflow enforcement system designed to optimize agentic behavior through a disciplined 'Plan-Do-Review' cycle. It leverages specialized artifacts like MAP.md for architectural context, automated journaling for history tracking, and 'anti-slop' policies to ensure high-quality, professional code output. By implementing a continuous improvement loop that captures and promotes lessons learned, Atris helps developers maintain organized repositories and ensures that Claude acts with unambiguous intent and consistent quality standards.
Key Features
01Context-aware MAP.md system for instant architectural alignment and reduced search tokens
02Self-improving feedback loop that captures and promotes development lessons to core policies
03Strict Plan-Do-Review execution cycle to prevent premature or incorrect coding
0454 GitHub stars
05Automated journaling and task management via integrated TODO.md and logs
06Anti-slop policy enforcement to maintain high-quality, professional coding standards
Use Cases
01Automating project logs and feature roadmap updates during active development cycles
02Managing complex coding tasks through structured intention capture and ASCII visualization
03Maintaining high-quality documentation and architectural context in large repositories
