Claude Opus 5 Benchmarked on SlopCodeBench

Benchmark results for Anthropic's Claude Opus 5 on SlopCodeBench — a test designed to evaluate AI coding agents — have been published on GitHub. The evaluation is part of a project on advanced context engineering for coding agents. The results aim to give a realistic picture of the model's coding capabilities.