Audit and rewrite evaluation/numeric claims to ensure they carry minimal protocol context (task + metric + constraint) and avoid underspecified model naming. Trigger: evaluation anchor checker, numeric claim hygiene, underspecified numbers, protocol context, 评测锚点检查, 数字断言, 指标上下文.
$npx -y skills add willoscar/research-units-pipeline-skills --skill evaluation-anchor-checkerInstalls into the current project.
Run `npx skills use "https://github.com/willoscar/research-units-pipeline-skills" --skill "willoscar/research-units-pipeline-skills/evaluation-anchor-checker"` and follow the generated skill instructions now. Read its complete output, redirecting it to a temporary file first if necessary. Resolve relative paths from the supporting-files directory it provides.
Use the skills in "https://github.com/willoscar/research-units-pipeline-skills" that are relevant to the current task. Run `npx skills add "https://github.com/willoscar/research-units-pipeline-skills"` and select the relevant skills, then follow their instructions.