DeepYardDeepYard
G

Glite ARF

Verifier-driven framework for parallel LLM coding agents with built-in audit capabilities

Open SourceFree

About

Academic research framework designed for reproducible multi-agent experiments with automated verification. Enables parallel execution of multiple LLM coding agents while maintaining rigorous audit trails and verification checkpoints. Built in Python for researchers studying agent coordination, correctness verification, and reproducible AI systems. Particularly suited for academic studies comparing agent architectures and verifying agent-generated code quality.

Details

Language
Patterns

Tags

frameworkmulti-agentcoding-agentpythonopen-sourceevaluation