guided-learning

Builds step-by-step learning paths with exercises, feedback, and resumable progress tracking.

1|Updated Jul 16, 2026
One-click install
npx skills add https://github.com/sota411/codex-config --skill guided-learning-sota411
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: guided-learning
Source: https://github.com/sota411/codex-config/tree/main/user-skills/guided-learning
Command: npx skills add https://github.com/sota411/codex-config --skill guided-learning-sota411

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? Self-directed learning often stalls because learners lack a structured path, get no feedback on their answers, and lose progress between sessions. This Skill turns any topic or study material into a guided, step-by-step learning journey with exercises, comprehension checks, and persistent progress that can be resumed later. ## Core Features & Use Cases - Structured Learning Paths: Converts a topic, goal, and source materials into a destination with a recommended route and only the next 2-3 steps visible, introducing one new concept per turn. - Active Learning Loop: Each turn requires the learner to recall, explain, apply, or create; feedback highlights strengths, isolates one misconception, and shows a model answer only after the learner responds. - Persistent Progress: A local Python script stores multiple learning tracks with revision-safe updates, history, archive/restore, export, and validated deletion, so sessions can resume exactly where they stopped. - Use Case: A learner pastes a textbook chapter on SQL joins and asks to learn it step by step. The Skill sets an observable goal, poses one retrieval question, adapts difficulty (support/standard/challenge) based on answers, and saves progress for the next session. ## Quick Start Use the guided-learning skill to teach me this topic step by step with exercises and save my progress so I can continue later.

Frequently Asked Questions about guided-learning

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I learn a topic step by step with an AI tutor?▼

Provide the topic, goal, and any study materials, and the Skill builds a destination with a recommended route and one learning action per turn. It alternates retrieval questions and new-concept exercises, giving feedback only after you answer.

How do I resume a learning session where I left off?▼

Ask to continue learning, and the Skill lists active tracks, loads the saved goal, current step, and next action, then resumes with a short recall question. Progress is stored locally under the Codex state directory via the progress_store.py script.

Does the progress store work without external services?▼

Yes, progress_store.py uses only the Python standard library and stores JSON tracks locally with file locking, atomic writes, and permission checks. No network access, accounts, or third-party packages are required.

What happens if two updates conflict or a write is interrupted?▼

Updates require an expected revision, so conflicting writes fail with a revision_conflict error instead of overwriting data. Interrupted mutations are recovered through a transaction journal, and corrupted or unknown-schema files are reported without being overwritten.

Can I delete or export my learning progress data?▼

Yes, tracks can be exported as JSON on request and deleted only after confirming with the exact track ID. You can also opt out of recording entirely and run the session as temporary with nothing saved.