Stats
Actions
Tags
How this command is triggered — by the user, by Claude, or both
Slash command
/cairn:benchThe summary Claude sees in its command listing — used to decide when to auto-load this command
Run `cairn bench` in the project root and report the results: the token savings from AST signature outlines, the re-read killer, and shell-output compression. Note that every compression is lossless.
Run cairn bench in the project root and report the results: the token savings from AST signature
outlines, the re-read killer, and shell-output compression. Note that every compression is lossless.
npx claudepluginhub vellixia/cairn --plugin cairn/benchRuns performance benchmarks for RLM-Claude-Code using pytest, including latency tests for key components. Supports comparisons, JSON output, and specific benchmarks.
/benchGenerates and runs table-driven Go benchmarks for a file, function, or package with profiling, baseline comparisons, and optimization suggestions.
/benchRuns typegraph-mcp benchmarks on the project, measuring token usage, latency, and accuracy. Reports token reductions, p50/p95/avg latencies per tool, scenario verdicts, and summary scores.