Help us improve
Share bugs, ideas, or general feedback.
From Bloom
Guides users to systematically learn a new field by collaboratively building a knowledge graph of concepts, purposes, and parent-child relationships. Best for uncertainty about where to start, feeling unsystematic, or wanting a full picture.
npx claudepluginhub li-evan/bloom --plugin bloomHow this skill is triggered — by the user, by Claude, or both
Slash command
/bloom:learn-graphThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
> 核心信条:**自己一步步建图谱的过程,本身就是最有效的学习——不要直接套用别人给的图谱。** 绝大部分知识,都有一个从常识就能入门的点。
Teaches any new concept through five structured perspectives: crossover (leverage existing knowledge), occam (scope depth), graph (knowledge map), prototype (minimal hands-on), and feynman (questioning).
Creates structured teaching sessions with mission, glossary, resources, and learning records for any topic.
Delivers structured multi-session tutoring for technical topics with Socratic diagnostics, knowledge graphs for prerequisites, agendas, teaching, quizzes, and progress tracking.
Share bugs, ideas, or general feedback.
核心信条:自己一步步建图谱的过程,本身就是最有效的学习——不要直接套用别人给的图谱。 绝大部分知识,都有一个从常识就能入门的点。
用户要系统进入一个新领域,或焦虑"学得不够系统 / 不知何时算够"。
用户为什么学 X?(接 learn-occam 的"既定问题")目的决定图谱画到多细。
概念/名称 · 用途 · 上下文关系(父子节点):
从入门点出发、沿父子关系排一条有效路径。颗粒度按需自由切换(领域图 → 细分学科图)。"学到哪算够"= 覆盖到能解决第一步那个目的的节点即可,不必学满。
learn-prototype(在图上找"最垃圾原型"的起点)。⚠️ 铁律·只用确证的已会知识:判断用户「已经会什么」只能用他确证学过的知识(亲口确认或可靠背景);严禁把「正在讲的材料 / 文章作者背景 / 对话里别人的知识」当成用户会的。拿不准 → 直接问「⚠️ 你学过 ___ 吗?」,绝不替他假设。
learn-occam(该不该学) learn-crossover(已会什么) learn-prototype(动手) learn-feynman(自查)。