new-learning-path

Prompt file imported from MohammadBafkar/Roadmaps (.windsurf/workflows/new-learning-path.md). Fill in {{arguments}} before use. Copyright stays with the author.

Create a curated cross-roadmap learning sequence with tests and exercises for a given topic

Mode: Content Creation — provide complete, production-ready examples. Do NOT use hint escalation.

Create a curated learning path for {{arguments}} that selects prerequisites and the most relevant roadmaps, orders them into a seamless progression, and includes tests and exercises at each stage.

Pre-conditions

• A learning path references existing roadmaps — it does not duplicate content.
• Check {{arguments}} for any user-specified focus areas, target role, or specific roadmaps to include.
• The output should be placed in a learning-paths/[topic]/ directory at the repo root.
• Each section must link to existing roadmap content and add bridging context.
• Every section must end with a test. A final comprehensive exam must close the path.

Process

Step 1: Research & Scope

1. Scan INDEX.md to identify all roadmaps relevant to {{arguments}}.
2. Identify prerequisite knowledge and which roadmaps cover it.
3. Determine the optimal ordering (dependency-based, not arbitrary).
4. Research latest industry expectations for the topic.

Step 2: Build Dependency Graph

Map which roadmaps depend on which:

Example for "Kubernetes":
  Linux/Unix Basics
         │
         ▼
    Networking ──────────┐
         │               │
         ▼               ▼
      Docker ────▶ Kubernetes
         │               │
         ▼               ▼
     Terraform     Service Mesh

Step 3: Create Structure

Follow the directory structure and templates defined in .ai/templates/learning-path.md.

Create:

1. README.md — Overview, dependency graph, time estimates, completion checklist
2. Section files (01-xxx.md, 02-xxx.md, ...) — Each section:
   • Links to specific roadmap files/levels (doesn't duplicate content)
   • Adds bridging context explaining how the previous section connects
   • Lists key topics to focus on
   • Points to the section test
3. Section tests (section-tests/01-xxx-test.md, ...) — Each test includes:
   • Multiple choice questions (5-10)
   • Short answer questions (3-5)
   • At least one practical challenge
   • All answers hidden behind <details> blocks
   • Score tracker
4. Diagnostic checks per section — 2-3 questions at the start of each section file, answered before study, to prime the brain and gauge starting knowledge. Follow the pattern in .ai/templates/learning-path.md.
5. Final exam (final-exam.md) — Comprehensive assessment:
   • Part 1: Concept review across all sections (30%)
   • Part 2: Practical application combining multiple sections (40%)
   • Part 3: Scenario-based problems with trade-off analysis (30%)

Step 4: Add Bridging Context

For each section transition, explain:

• What the learner just mastered
• Why the next topic builds on it
• Specific connections (e.g., "You learned about TCP in networking — K8s Services use TCP/UDP for pod-to-pod communication")

Step 5: Define Time Estimates

Provide three paths:

Step 6: Spaced Review Calendar

Include a spaced review recommendation in the README:

• After each section: register key topics in _progress/review-state.json
• After completing the path: run generate-review weekly to maintain retention
• Section test scores feed into spaced review scheduling

Step 7: Structural Consistency

• Update INDEX.md if the learning-paths directory is new.
• Ensure all links to roadmap files are valid.
• Follow .ai/guidelines/markdown.md formatting.