BenchBox is a "benchmarking toolbox" that makes it simple to benchmark analyitic (OLAP) data tools with SQL and dataframes.
Latest release v0.4.2 — BenchBox 0.4.2 · 5 Oct 2026
16 files for Claude Code, Codex, OpenCode and Gemini CLI: blind-spot, pr, benchbox, benchbox and 12 more — 2,660 tokens loaded in every session.
AGENTS.md A 2,598 tok CLAUDE.md A 33 tok GEMINI.md A 29 tok .claude/settings.json A — .claude/skills/benchbox/SKILL.md A 68 tok .claude/skills/code/SKILL.md A 129 tok .claude/skills/docs/SKILL.md A 40 tok .claude/skills/shared-agent-execution/SKILL.md A 62 tok .claude/skills/shared-change-framework/SKILL.md A 34 tok .claude/skills/shared-investigation-framework/SKILL.md A 31 tok .claude/skills/shared-review-protocol/SKILL.md A 42 tok .claude/skills/skill-sync/SKILL.md A 23 tok .claude/skills/tidy-perms/SKILL.md B 42 tok .claude/skills/todo/SKILL.md A 186 tok .claude/commands/blind-spot.md A 28 tok .claude/commands/pr.md A 20 tok These files are BenchBox-dev/BenchBox's own configuration — they tell Claude Code, Codex, OpenCode and Gemini CLI 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.