Learning Path Agent

An intelligent agent that designs personalized study plans for any topic you want to learn. It curates resources from real platforms, sets achievable milestones, builds assessments to verify understanding, and adapts difficulty based on your progress. Whether you are learning a new programming language, preparing for a certification, or diving into a completely new field, this agent structures your journey from beginner to proficient.

Supported Platforms & Integrations

Platform	Integration Type	Features
Coursera	Resource Curation	Links to relevant courses, specializations, and guided projects with difficulty ratings
YouTube	Video Resources	Curates playlists and standalone tutorials organized by topic and skill level
Obsidian	Note Sync	Exports study plans as linked markdown notes with backlinks for your vault
Anki	Spaced Repetition	Generates flashcard decks aligned with each learning milestone
Khan Academy	Foundation Learning	Maps prerequisite knowledge and links to foundational lessons
GitHub	Project-Based Learning	Links to repositories, coding challenges, and open-source projects for hands-on practice

When to Use

• Learning a new programming language or framework — generates a structured path from syntax basics through advanced patterns with hands-on projects at each stage
• Preparing for professional certifications — maps exam objectives to study materials, practice tests, and a timeline that fits your schedule
• Career pivot into a new field — builds a comprehensive curriculum covering foundational knowledge, practical skills, and portfolio-building projects
• Deep-diving into an academic subject — creates reading lists, lecture sequences, and comprehension checkpoints for topics like machine learning, economics, or philosophy
• Teaching or mentoring someone else — generates a structured plan you can hand to a mentee with clear expectations and milestones
• Brushing up on forgotten skills — identifies knowledge gaps through diagnostic questions and creates a targeted refresher plan

Alternatives to Consider

• Use a simple checklist if you already know exactly what resources to study and just need to track completion
• Use Coursera or Udemy directly if you prefer a single platform's guided learning experience over a multi-source approach
• Use a study group or bootcamp if you need live accountability and peer interaction rather than self-paced study

Quick Start

Copy
agent: learning-path-agent
config:
  topic: "Rust programming language"
  current_level: "beginner"
  target_level: "intermediate"
  weekly_hours: 10
  preferred_formats:
    - video
    - hands-on
    - reading
  deadline: "2026-06-15"
  include_assessments: true

Example Prompt

Create a learning path for Rust programming. I have experience with Python 
and JavaScript but have never used a systems language. I can dedicate 
10 hours per week and want to be comfortable building CLI tools and 
small web services within 3 months.

Example Output

## Learning Path: Rust Programming (Beginner → Intermediate)
Timeline: 12 weeks | 10 hrs/week | Total: ~120 hours

### Phase 1: Foundations (Weeks 1-3)
Milestone: Write and compile basic Rust programs with ownership understood
- [The Rust Book, Ch 1-8](https://doc.rust-lang.org/book/) (8 hrs)
- [Rustlings exercises](https://github.com/rust-lang/rustlings) (10 hrs)
- [YouTube: Let's Get Rusty playlist](https://youtube.com/...) (6 hrs)
- Assessment: Build a temperature converter CLI tool

### Phase 2: Intermediate Concepts (Weeks 4-7)
Milestone: Handle error patterns, traits, and generics confidently
- [The Rust Book, Ch 9-15](https://doc.rust-lang.org/book/) (10 hrs)
- [Exercism Rust Track](https://exercism.org/tracks/rust) (14 hrs)
- Assessment: Build a file search utility with error handling
...

Advanced Configuration

Platform-Specific Settings

Copy
platform_preferences:
  coursera:
    include_paid: false
    prefer_university_courses: true
  youtube:
    max_video_length: 30  # minutes
    prefer_channels:
      - "Let's Get Rusty"
      - "Jon Gjengset"
  anki:
    cards_per_milestone: 25
    export_format: "apkg"

Parameters Reference

Parameter	Type	Default	Description
topic	string	required	The subject or skill to learn
current_level	string	"beginner"	Starting level: beginner, elementary, intermediate, advanced
target_level	string	"intermediate"	Goal level to reach by deadline
weekly_hours	number	8	Hours available for study per week
deadline	date	null	Target completion date; null for open-ended
preferred_formats	array	["video", "reading"]	Content formats: video, reading, hands-on, audio, interactive
include_assessments	boolean	true	Generate checkpoints and mini-projects per milestone
assessment_style	string	"project"	Assessment type: quiz, project, essay, mixed
difficulty_curve	string	"gradual"	Progression speed: gentle, gradual, aggressive, custom
prerequisite_check	boolean	true	Verify foundational knowledge before starting
resource_budget	string	"free"	Resource cost filter: free, low (under $50), any
language	string	"en"	Preferred language for resources

Core Concepts

Concept	Description
Milestone Mapping	Each phase ends with a concrete deliverable that proves comprehension before advancing
Adaptive Pacing	If assessments show struggle, the agent inserts remedial resources; if you breeze through, it accelerates
Resource Triangulation	Every concept is covered by at least two different formats to reinforce understanding from multiple angles
Spaced Review	Key concepts resurface in later milestones through flashcards and callback exercises
Prerequisite Graphs	Topics are ordered based on dependency analysis so you never encounter a concept before its foundation

┌─────────────┐     ┌──────────────┐     ┌─────────────┐
│  Diagnostic  │────▶│ Phase Design  │────▶│  Resource    │
│  Assessment  │     │ & Milestones  │     │  Curation    │
└─────────────┘     └──────────────┘     └──────┬──────┘
                                                 │
┌─────────────┐     ┌──────────────┐             │
│  Adaptive    │◀───│  Progress     │◀────────────┘
│  Adjustment  │    │  Tracking     │
└─────────────┘     └──────────────┘

Workflow Examples

Scenario 1: Career Pivot to Data Science

Input:

I'm a frontend developer wanting to transition to data science. 
I know basic statistics from college. Budget: $100 for courses. 
Timeline: 6 months at 15 hours/week.

Output: A 24-week plan covering Python for data science, statistics refresh, pandas/numpy, machine learning fundamentals, and a capstone portfolio project. Resources pulled from Coursera (Andrew Ng's ML course), Kaggle competitions, and StatQuest YouTube channel. Monthly assessments include Kaggle notebook submissions.

Scenario 2: Language Learning (Japanese N3)

Input:

I passed JLPT N4 last year. Target: N3 by December. 
2 hours daily. Prefer audio and flashcards for commute time.

Output: A structured plan with Anki decks for N3 vocabulary and kanji, curated podcast episodes from JapanesePod101, grammar lessons from Tae Kim's guide, weekly reading assignments from NHK News Easy, and monthly mock exams.

Scenario 3: Professional Certification (AWS Solutions Architect)

Input:

Preparing for AWS Solutions Architect Associate exam. 
I use AWS at work for basic EC2 and S3. Exam date: May 20.

Output: A 10-week plan mapping all exam domains to study resources. Includes Stephane Maarek's Udemy course, AWS free tier lab exercises, Tutorials Dojo practice exams, and a weekly domain-focused study schedule with practice test benchmarks.

Best Practices

1. Complete the diagnostic honestly — Skipping prerequisites or inflating your level leads to frustration later when the plan assumes knowledge you do not have. The agent calibrates everything from this starting point.

2. Follow the milestone order — The prerequisite graph exists for a reason. Jumping ahead to interesting topics without completing foundations creates gaps that compound over time.

3. Report struggles early — Tell the agent when a milestone felt too hard or too easy. The adaptive system works best with honest feedback rather than silent completion.

4. Mix your resource formats — Even if you prefer video, engage with at least one hands-on project per phase. Research shows interleaving formats improves retention significantly.

5. Review Anki cards daily — If using the spaced repetition integration, consistency matters more than volume. Ten minutes daily outperforms an hour once a week for long-term retention.

Common Issues

Plan feels too slow or too fast

Adjust the difficulty_curve parameter. Switch from "gradual" to "aggressive" if you are consistently finishing milestones ahead of schedule, or to "gentle" if you are falling behind. You can also update weekly_hours to reflect your actual availability.

Resources are paywalled or unavailable

Set resource_budget to "free" to exclude paid content entirely. The agent prioritizes free resources by default but may include highly-rated paid courses if the budget allows. You can also specify preferred_platforms to limit sources to platforms you already have access to.

Learning path for a niche topic has few resources

For obscure or very new topics, the agent may not find enough structured courses. In this case, it falls back to primary documentation, research papers, and community forums. Add allow_research_papers: true and include_community_resources: true to expand the resource pool.

Privacy & Data Handling

All learning path data is stored locally in your project directory. Assessment results, progress notes, and study schedules remain on your machine. When the agent curates resources, it references publicly available course catalogs and does not transmit your learning history to any external service. If you enable Obsidian or Anki integration, files are written to your local vault or Anki media folder. No personal data, assessment performance, or study habits are shared with third parties. You can delete all generated plans and progress files at any time by removing the output directory.