Stet is a web-based tool for developers to evaluate and compare AI coding agents on real repository tasks. It replays actual work, scores agent changes, and helps teams make evidence-based decisions about deploying new models or configurations. The platform is designed for engineering teams seeking to improve code quality and agent reliability.
Latest indexed changes and source events
I used autoresearch to improve my AGENTS.md, measured against real tasks discovered by the PulseGate indexer
A brief investigation into the GPT-5.5 regression claims discovered by the PulseGate indexer
Other apps tracked under the same category.
