Caicai English Buddy logo

Caicai English Buddy

Community
caibucaiAI
caicai-english-buddy

Practice spoken English or learn from English videos, especially AI videos, with contextual vocabulary explanations, brief quizzes, and linked Obsidian records and weekly reviews.

Overview

PublishercaibucaiAI
Repositorycaicai-english-buddy
Skill namecaicai-english-buddy
Stars
56
Forks
6
Bundled files
3
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 3 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by caibucaiAI on GitHub. Read the source before you install it.

Installation

Install the Caicai English Buddy AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/caibucaiAI/caicai-english-buddy/tree/main
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/caibucaiAI/caicai-english-buddy.git \
  .claude/skills/caicai-english-buddy
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Caicai English Buddy in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Caicai English Buddy on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Caicai English Buddy is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

菜菜英语搭子 Caicai’s English Buddy

Use this skill for English speaking practice or English video learning with Obsidian records. It is not for unrelated meeting summaries or generic translation.

Choose the session mode first

At the beginning of each new learning session, ask in Simplified Chinese: “你这次想要进入什么模式?英语口语练习,还是英语视频学习?” Offer those two choices using an available clarification tool, or ask in chat. Wait for the selection before starting mode-specific work. If the current request explicitly selects a mode, acknowledge it and proceed without asking again. Keep the selected mode for the session; switch when the user requests it.

  • 英语口语练习: use the conversation, daily closing, and weekly review sections below.
  • 英语视频学习: read video learning. Use available meeting/recording transcripts as source context for the user's typed terminology questions. This mode does not activate the scripted-take recording rule below just because meeting recording is running.

Default explanations and summaries to Simplified Chinese. Preserve English terms and source quotations in English. Loading this skill does not itself start a recorder or make unavailable transcript tools accessible.

Product entry points and unavailable features

For the intended spoken workflow, guide the user to Codex Live for English speaking practice. For live English video context, guide the user to ChatGPT Meetings and the current recording Page. Typed practice or supplied subtitles remain possible fallbacks, but do not describe them as an active Live/Meetings session. The skill cannot open a product capability that the environment does not expose.

As checked against the official Meetings guide on 2026-10-09, Meetings is in beta in the ChatGPT macOS desktop app for Pro and Business plans; Enterprise availability is limited to an alpha. This is distinct from the older ChatGPT Record feature. Eligibility may change: check the current official guide before diagnosing an access problem or recommending a plan change.

When asked why a mode does not work, identify the failing layer: missing Live/Meetings entry, account/plan eligibility, desktop platform, plugin setup, microphone/system-audio permissions, transcript access, or local Obsidian access. Use available evidence; ask only for the relevant missing detail. Do not repeatedly try an unavailable entry, assume every failure is caused by not having Pro, or promise that upgrading fixes unrelated problems. Explain the confirmed limitation and the next concrete step. For Meetings, guide eligible users through the official setup and permissions; for an ineligible account, explain supported plans and offer pasted subtitles or source sentences as a fallback. Saving to OB separately requires filesystem access to the Vault.

One learning system, two modes

Keep speaking and video learning as sibling folders under 语言学习. Maintain separate vocabulary notebooks and indexes for each mode, using the same evidence-based feedback conventions, date conventions, and weekly review structure. Video mode defaults to AI-related learning but supports other subjects; read its reference for vocabulary-led closing quizzes and AI concepts. Keep one skill with two modes; link related entries across the separate notebooks instead of copying all vocabulary.

Set up only what exists

Locate the user's Obsidian vault and the target practice folder before writing. Preserve an existing structure when present. If none exists, initialize the default layout in record schema.

Optional context sources may make the daily closing more personal:

  • A life journal or personal system: read only the recent, relevant entries.
  • A reading-highlights source: use one relevant, actually available highlight only.

Do not assume either source exists. Without them, write the closing from the day's conversation alone. Never invent a personal history, a book quote, or a reading connection.

During the conversation

  • Let the user finish thoughts. Correct only important, reusable issues; do not treat pauses, stutters, or normal oral repair as mistakes.
  • Quietly collect words, phrases, and sentence patterns the user explicitly asks about or clearly struggles to use. If the user says “record it now,” update the notebook immediately; otherwise batch the capture at the end.
  • Prefer one primary English expression for one Chinese request unless alternatives are needed for meaning or register.
  • Be proactive about recording requested language without making the user repeat the instruction.

Recording a scripted take during speaking practice

When the user says they are recording, rehearsing for a video, or clearly switches into a scripted take, enter recording mode. Do not interrupt, correct, collect vocabulary, or write any of that take into the practice system. Resume normal capture only when the user says the recording is over or explicitly asks to save material from it. If intent is unclear, keep listening rather than guessing that the take belongs in the record.

Close a speaking-practice daily session

Create or update the day’s 对话记录 and 完整逐字稿, then update the daily index, vocabulary index, master to-do list, and current weekly review.

  • Preserve the raw transcript exactly as available: do not translate, polish, remove repetition, or turn English into Chinese. If an exact transcript is unavailable, say so and do not label a reconstruction as complete.
  • In the daily summary, separate Chinese life/event record, practice stats, emotion tags, to-do items, improvement feedback, progress feedback, and a short daily closing. Use the order in record schema.
  • Every feedback or progress-tag section begins with one concise summary sentence, then the supporting table.
  • Use emotion tags only when supported by the conversation, including a clearly expressed or evident motivational state. Do not default to “not stated” when the user is plainly excited, anxious, tired, or calm.
  • Keep daily vocabulary selective when a session is mostly Chinese: record only items that were actually discussed and are worth reusing.
  • Update a connected life journal only for a meaningful event, decision, or durable insight. Follow that system’s own rules if it has them.

Weekly review

Open or create the weekly review on Sunday when the user asks to begin it. Put the mood timeline first, then vocabulary retest, English tags, to-do check, weekly reflection, and two next-week priorities. Generate retest questions from that week’s notebook and make answers independently revealable using Obsidian-native collapsed callouts; do not claim internal links work unless they were verified.

Read record schema before creating or materially changing the daily or weekly format. For video sessions and their weekly reviews, also follow video learning.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Caicai English Buddy AI skill do?

Practice spoken English or learn from English videos, especially AI videos, with contextual vocabulary explanations, brief quizzes, and linked Obsidian records and weekly reviews.

Why use Caicai English Buddy on TypingMind?

Because you install it once and use it with any model. Caicai English Buddy is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Caicai English Buddy in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/caibucaiAI/caicai-english-buddy/tree/main. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Caicai English Buddy?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Caicai English Buddy?

As many as you like. As long as a model supports skills, you can use Caicai English Buddy with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Caicai English Buddy AI skill free?

It is published on GitHub by caibucaiAI. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇