AI Learning Coach plugin
A Codex and Claude Code plugin where a coach teaches on real projects, guides debugging and records evidence of mastery
- 230 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- The helper state tool runs through `uv` and writes learning archive files to your disk.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add plugins/ai-learning-coachDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The helper state tool runs through `uv` and writes learning archive files to your disk.
This is third-party code. Review the repository files before installing.
What it does
AI Learning Coach turns the agent into a personal coach for learning on real code. It targets developers who know Java or another backend language and want to move into TypeScript and agent development. The coach first reads the project and then offers a slice you can work on right now: predict, run, debug, change. The plugin has two skills. `learning-coach` leads daily learning and evidence recording: when you start, continue, did not understand a line or are debugging. `focus-coach` helps choose a stage and a main line when there are too many directions. Progress lives in a Learning Lab, your own Markdown files: goal, evidence, current state and stable knowledge. A helper tool launched through `uv` keeps the state: it records evidence, reminds about review, builds a weekly recap and computes percentages. When evidence is insufficient it refuses to raise a level. Without `uv` ordinary skill-based teaching continues.
Who it is for. Developers with Java or another backend language background who learn TypeScript, Python or Go on real repositories with Codex or Claude Code.
Good fit when
- You want to start learning TypeScript but it is unclear from which file in an agent repository.
- You answered right yesterday and cannot explain today and need a comprehension check.
- You need to come back to study after a few days and continue from the right point.
- There are too many directions and you must decide what to learn first.
Not a fit when
- You need a ready course with lectures and grades: the coach works around your code and questions.
- You need per-person statistics for a school: data lives in your own Markdown files.
- You need English material: texts and the coach are oriented to Chinese by default.
Example request
I know Java and want to learn TypeScript on a real agent project. Pick the first file and start me on one small step.Limitations
Materials for studying source code are the strongest, other areas such as writing, languages and exams are covered by examples. Work runs in Chinese by default and another language can be requested explicitly. The authors say first-time completion by real people and seven-day reuse still await feedback and no measurements exist. State is kept by a tool through `uv`, without it ordinary teaching remains with no automatic progress calculation. For Claude Code install goes through the marketplace, for Cursor and Gemini CLI a skills-only install is available, details in `docs/install.md`. The README is in Chinese with an English version in `README.en.md`.
How to disable. Remove the plugin with `claude plugin uninstall ai-learning-coach@ai-learning-coach` or through Codex tools. Learning Lab files stay in your folder as Markdown.
FAQ
How do I install the plugin in Codex?
With `codex plugin marketplace add chrichuang218/ai-learning-coach` and `codex plugin add ai-learning-coach@ai-learning-coach`, then start a new session.
How do I install it in Claude Code?
With `claude plugin marketplace add chrichuang218/ai-learning-coach` and `claude plugin install ai-learning-coach@ai-learning-coach`.
Is uv needed?
It is needed for the state tool: evidence records, reviews and progress calculation. Without it skill-based teaching works but with no automatic calculation.
Where is progress stored?
In the Learning Lab, your own Markdown files: goal, evidence, state and stable knowledge. Obsidian reads them too.
What if I have no project?
The coach picks a real repository from the starter tracks: a TypeScript agent, Python, Go, frontend or Rust.
Discussion
No questions or reviews yet.
Related
PluginsFive skills turn Claude Code or Codex into a tutor: a knowledge probe, a topic map, one step of teaching and a lock-in quiz.
A Codex plugin for daily knowledge training: the AI asks questions, your own server scores answers and schedules reviews.
A personal AI tutor for Claude Code: a syllabus, one lesson at a time, and each next lesson shaped by your annotations and feedback
A tutor for any subject: diagnosis, a study plan, project tasks and mastery checks, with the results saved to Markdown files
Study coach plugin for people with ADHD: small chunks of new material, review quizzes and explain-it-back prompts.
Claude Code plugin that builds a study plan for any topic, runs adaptive quizzes and schedules reviews with the SM-2 algorithm.
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Repository stars | Main difference |
|---|---|---|---|
| AI Learning Coach (this card) | no | 230 | |
| tutor-skills | no | 1,402 | tutor-skills gives tutor skills, while AI Learning Coach teaches on your real code and computes progress from recorded evidence. |
| Good Learning | no | 326 | Good Learning offers learning techniques, while here there are two coaches and a state tool for reviews and recaps. |
README in short
The README explains how the coach works, how to install the plugin for Codex and Claude Code, how the Learning Lab is built and which checks exist. MIT license, with an English version.