Dependency Graph Prompt

Prompt

You are an expert curriculum designer and graph theorist.

Below is an ordered list of concepts from a college-level course. For each concept, identify which other concepts from the list are direct prerequisites — that is, concepts a student must understand before they can learn this one.

Return the result as a CSV with the following columns: - concept_id (integer, 1-based) - concept_name (string) - dependencies (pipe-separated list of concept_ids that are prerequisites, or empty if none)

Rules: - Only list direct prerequisites, not transitive ones. - Foundational concepts (id 1–20) should have few or no dependencies. - Every concept should be reachable from at least one foundational concept. - Avoid circular dependencies.

Concept list: [PASTE NUMBERED CONCEPT LIST HERE]

Parameters Used

• Course: Inverting the Impossible: Systematic Thinking for Innovation Radiation
• Concepts: 200 (IDs 1–200)
• Output format: CSV with columns: concept_id, concept_name, dependencies
• Model: Claude Sonnet
• Skill version: 0.05

Notes

The dependency graph generated by this prompt is stored as:

• learning-graph.csv — raw CSV output
• learning-graph.json — converted to JSON for the graph viewer

The graph was validated with validate-learning-graph.py to confirm no circular dependencies and that all 200 concepts are reachable from foundational nodes. See Graph Quality Analysis for the full validation report.