Inspired by SWE-Bench, for Business Central (AL) ecosystem.
BC-Bench is a benchmark for measuring coding agents on real-world Microsoft Business Central development tasks written in AL. It helps compare models and agent configurations using reproducible evaluation data.
Latest release v0.12.0 · 14 Sept 2026
3 files for GitHub Copilot, Claude Code and Codex: BC-Bench copilot-instructions.md, create-release, update-agent-harness — 996 tokens loaded in every session.
.github/copilot-instructions.md A 996 tok .github/skills/create-release/SKILL.md A 56 tok .github/skills/update-agent-harness/SKILL.md A 58 tok These files are microsoft/BC-Bench's own configuration — they tell GitHub Copilot, Claude Code and Codex how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.