用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt;运行前请先检查来源。
npx skills add https://github.com/imbue-ai/catalyst --skill score-length命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地?可下载 SkillsMP 当前能够提供的文件。
基于 SOC 职业分类
正在显示 SKILL.md
| name | score-length |
| description | Score the length of a given theory. |
| argument-hint | the theory ID to score (e.g. T_20260414_143100_d4e5f6). Optionally, a STANDARD_LENGTH value. |
| user-invocable | false |
This skill explains how to assess and score the length of a given theory based on its content.
Besides the theory to score, you may optionally be given a STANDARD_LENGTH value. The default STANDARD_LENGTH, if not specified, is 7000 words.
theory.md in number of words, excluding any conclusion sections and appendices, as well as any introductory sections (such as phenomenon descriptions, overviews and summary sections, statement/content tables, prior art comparisons, etc):theory.md file (e.g. grep -n '^#' <CONTEXT_DIR>/theory/theory.md).head -n 500 <CONTEXT_DIR>/theory/theory.md | tail -n +50 | wc -w).Length Score = 1 / (1 + (words_in_main_body / STANDARD_LENGTH)**3), where words_in_main_body is the number of words in the main body of the theory. Round to two decimal places. Use ad-hoc Python code to perform this calculation, as described above.