Guide · Agent skill
How to use ljg-explain: teach concepts until they are usable
Updated · written with AI from the project's own documentation and checked against our catalogue data
ljg-explain is an agent skill, a set of written instructions that changes how Claude explains an idea. Instead of giving a definition and stopping, it teaches a concept, formula or mechanism so you can recognise new examples, explain why a result happens and adjust your judgement when a condition changes. This guide covers what the skill does, how to install it, how to trigger it with natural requests and what to watch for.
ljg-explain: Teach Concepts Until Usablelijigang/ljg-skills · MIT Download the skill Quick start
- 1Download the ZIP. Download the ljg-explain skill ZIP from this page.
- 2Unzip the skill folder. Unzip the archive so the ljg-explain folder containing SKILL.md is ready to copy.
- 3Place it in a skills folder. Put the folder in ~/.claude/skills/ for personal use, or in .claude/skills/ inside a project.
- 4Or upload it in the Claude apps. In the Claude apps, open Settings, go to Capabilities, then Skills, and upload the ZIP.
- 5Ask for an explanation. Request an explanation of a concept or formula, naming ljg-explain or using a phrase such as 讲透这个概念或公式.
What does the ljg-explain skill make the agent do?
The skill sets a completion standard for an explanation: the reader should understand why the idea is needed, be able to identify it, explain how the result comes about, see how the answer changes when one condition changes, and apply the idea to a different situation.
It starts from a real question rather than a bare definition. It favours examples and close contrasts where the deciding feature is clear, walks through the causes behind a result, and in formula explanations ties each symbol to a concrete quantity and unit. Chinese explanations are written in plain, conversational language.
- Connects the concept to a concrete situation first
- Shows typical examples and near-miss contrasts with the reasons for the difference
- Explains the cause-and-effect process or the definitions and quantity relationships
- Changes one key condition and shows the effect
- Applies the idea to a different situation at the end of the explanation
How do you install the ljg-explain skill?
Download the skill ZIP from this page and unzip it. Then place the skill folder in one of two locations: your personal skills folder at ~/.claude/skills/, or a project-specific folder at .claude/skills/ inside the project you are working in.
If you use the Claude apps instead, open Settings, go to Capabilities, then Skills, and upload the ZIP there.
- Personal use across projects: ~/.claude/skills/
- One project only: .claude/skills/ inside that project
- Claude apps: Settings → Capabilities → Skills, then upload the ZIP
~/.claude/skills/ljg-explain/SKILL.mdHow do you trigger ljg-explain with a natural request?
You can call the skill by name, ljg-explain, or simply ask for an explanation in the wording the skill is written to respond to. Its description lists phrases such as 通俗易懂地讲解 (explain in plain, easy-to-follow language), 讲透这个概念或公式 (explain this concept or formula thoroughly), 费曼讲解 (Feynman-style explanation) and ELI5. It also fits requests where you want to understand a point from scratch and use it in a new situation.
Ask about a single concept, formula or mechanism. Say what you already know or where you get stuck if you can, because the skill adjusts the level of the explanation to your background.
- 用ljg-explain讲解一下似然 (Use ljg-explain to explain likelihood)
- 请通俗易懂地讲解方差公式 (Please explain the variance formula in plain language)
- 讲透正反馈原理 (Explain the principle of positive feedback thoroughly)
- I already know basic probability. Explain Bayes' rule with ljg-explain so I can apply it to a new case.
What does a good result look like?
The explanation is delivered in one complete response in the chat, and it ends where the explanation naturally finishes. It does not add a quiz or answer key unless you ask for an interactive session. By default it creates no notes or files; if you ask it to save the output, it follows the format and location you specify.
For a formula, expect the symbols mapped to real quantities, a simple numerical example, and a note on when the formula does and does not apply. For a mechanism, expect the chain of cause and effect and the conditions under which it holds.
- One complete explanation in the chat, not a series of quizzes
- No notes, cards or reminders created unless you request a file
- Formulas: symbols tied to units, a worked small example, and stated assumptions
- Mechanisms: the feedback or causal chain, limits, and an application to another case
What does the skill need, and is it safe to use?
The skill lists no requirements. It runs no scripts and makes no network calls, according to its own description. It consists of written instructions for the agent, and it is designed to be self-contained: it does not call another explanation skill or create other outputs as part of a teaching task.
Because it runs no code, the main thing to check is the content of the answer itself. The skill asks the agent to separate facts, assumptions and simplifications, and to say when a number or example is hypothetical. Verify any real-world data against a primary source before you rely on it.
- Requirements: none listed
- Runs scripts: no
- Uses network: no
- Licence: MIT
What tips help you get better explanations?
State the concept precisely and mention the context you need it for, such as a statistics exam or a product decision. Ask for one idea at a time, since the skill is built to explain a single point thoroughly.
If an explanation is not clear, ask the agent to change one condition in the example, or to give a case that looks different on the surface but depends on the same key relationship.
- Name the exact formula or mechanism
- Give your level of background if you know it
- Ask for a changed condition to see how the result shifts
- Ask for a new situation to test transfer
What people use it for
- Understanding a statistics concept such as likelihood well enough to use it on a new dataset
- Preparing to explain a maths or science formula to students in plain Chinese
- Learning how a feedback loop or other mechanism works, including when it stops working
- Getting a conversational explanation of a principle before reading a technical source
Frequently asked questions
How do I install the ljg-explain skill?
Download the ZIP from this page and unzip it. Put the folder in ~/.claude/skills/ for personal use or in .claude/skills/ inside a project, or upload the ZIP in the Claude apps under Settings, Capabilities, Skills.
Is ljg-explain free to use?
The skill is released under the MIT licence. This page does not list any price for it, so check the page you downloaded it from for any terms that apply to your use.
Does ljg-explain need any other software or accounts?
The skill lists no requirements, runs no scripts and uses no network. It is a set of written instructions for the agent.
What is the difference between ljg-explain and a simple summary?
A summary condenses content. This skill is built to teach an idea so you can recognise examples, explain results and apply it to new cases. It is not meant for summarising a whole book or paper.
Can I use ljg-explain to proofread or translate?
No. The skill's own description says it is not for simple polishing or translation, nor for reading companions, whole books or papers, or for formal proofs or step-by-step procedures alone.
Why doesn't the skill give me a quiz after the explanation?
By default it delivers one complete explanation and stops where the explanation naturally ends. It only runs in an interactive, multi-turn way if you explicitly ask for that.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
Use ljg-explain if you want a clear, conversational Chinese explanation that helps you use a concept in a new situation, especially for statistics, maths and science. Pick something else if you need proofreading, translation, a summary of a book or paper, or only a formal proof or procedure without intuition.
Download the skill