Benchmark
Run the laconic benchmark and print results in the CLI. Use when the user invokes /laconic benchmark or asks to benchmark token savings.From its SKILL.md
npx -y skills add brillcp/Laconic --skill benchmarkAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
1.4 KB, as published. Nobody here has run it
Run the laconic benchmark from anywhere — locate the script in the installed plugin and execute it.
Steps
-
Locate the plugin install path. Try in order:
$CLAUDE_PLUGIN_ROOT(set when this skill runs)~/.claude/plugins/cache/laconic/laconic/*(newest match)- The current repo if
pwdis the laconic repo Fail with a clear message ifbenchmark/run.shcannot be found.
-
Check prerequisites:
claudeandjqonPATH. If missing, tell the user how to install. -
Run the benchmark via Bash. Honor an optional
RUNSargument from the user (default 3). Stream stderr so the user sees per-prompt progress. The script writesbenchmark/results.mdinside the plugin path. -
Print results to chat: read
benchmark/results.mdand show the full markdown table. Include the total saved % on the first line of the reply.
Rules
- Don't truncate the results table — user wants the full picture.
- If a run errors out (no
claudeCLI, nojq, no prompts file), stop and report — don't fake numbers. - Long-running. Warn upfront if
RUNS× number of prompts × 2 modes will take more than a couple of minutes. - Don't push or commit changes to
results.md— that's the user's call.