EDCA-OS (Expression-Driven Cognitive Architecture) is a newly released framework that proposes a protocol-governed, expression-driven approach for controlling LLM behavior.
The public v0.1 release includes:
• EDCA Protocol Stack (L0–L5)
• Semantic Control Layer (public-safe)
• Yuer DSL – a language designed for deterministic LLM execution
• Behavioral Evidence Pack v1.0
• High-level architecture map (public)
It focuses on deterministic behavior, semantic-layer safety, and expression-to-protocol-to-behavior workflows.
No runtime implementations or proprietary logic are released.
Only conceptual architecture and public-safe behavioral evidence.
I’m the author.
Happy to discuss the architecture or the expression-driven model.
All internal mechanisms remain closed-source; this release focuses on conceptual and behavioral evidence only.
yuer2025•1h ago
The public v0.1 release includes:
• EDCA Protocol Stack (L0–L5) • Semantic Control Layer (public-safe) • Yuer DSL – a language designed for deterministic LLM execution • Behavioral Evidence Pack v1.0 • High-level architecture map (public)
It focuses on deterministic behavior, semantic-layer safety, and expression-to-protocol-to-behavior workflows.
No runtime implementations or proprietary logic are released. Only conceptual architecture and public-safe behavioral evidence.
GitHub: https://github.com/yuer-dsl/EDCA-OS