kenhia/harness-eval: Comparative evaluation of AI coding-agent harnesses (POC run 1): same model, same prompt, five harness setups, graded by independent AI reviewers

Source-linked topic cluster with 1 signals across related articles, projects, models, papers, and source updates.