Multi-agent RISC-V verification and test-generation framework for AI-assisted RTL, ISS, compliance, coverage, and debug workflows.
HEALTH ? 98 / Healthy
// momentum
// why it's here
- Active: last commit today
- Built almost entirely by one person
- Health 98/100 with only 12 stars
// how to help
// readme
AVA β Autonomic Verification Agent
A multi-agent RISC-V RTL verification platform, and a research track studying whether verification agents can be trusted.
Quick Start Β· Architecture Β· VOE Research Β· Reproducing Results Β· Contributing
The verdict
completed/ exit 0: verified.failed/ exit 1: a violation was found.incomplete/ exit 3: it could not verify (a tool was missing, or a stage could not run).
The exit code is the verdict, and it does not depend on the console encoding. Current release: v4.0.1. It supersedes v4.0.0, whose completed compared only instruction PCs on real Spike logs.