systems-thinking-skills

Commission an interactive stock-flow simulation web app from your AI coding agent (Claude Code or Codex). Three phases: readback with forced ≥1 disagreement, brief emission, hand-computed extraction test. Pass 1 enforces hard caps (1 stock / 2 flows / 2 params / ≤30 steps / linear only); Pass 2 extends. Optional Phase 0 interview mode for participants without a diagram.

6 skills

Critique a hand-drawn stock-flow / causal-loop diagram, identify the system archetype (Limits to Growth, Shifting the Burden, Fixes that Fail), and surface Meadows leverage points. Refuses to draw the diagram for you — drawing is the act of thinking.

6 skills

Find the ONE binding constraint in a flow/process system using Goldratt's Theory of Constraints, classify it as a resource or (usually) a policy, and return the 5 Focusing Steps as concrete actions — Exploit and Subordinate before Elevate. Refuses to optimize non-constraints, to Elevate before Exploit/Subordinate, to proceed without a goal, or to force a single constraint on a demand-constrained or exploratory system.

6 skills

Map elements of a stock-flow model to Donella Meadows' 12 leverage points, ranked by potential impact for a stated goal. Refuses to invent stocks, flows, or parameters not in the model. Refuses to return a ranking without an explicit, measurable goal. Returns exactly 3 candidates with at least one at knob ≤5 (the asymmetry lesson).

6 skills

Take a trade-off ('improving X worsens Y') and dissolve it with TRIZ — act as the contradiction matrix over the full 40 inventive principles + separations + resource analysis, while enforcing the discipline the LLM skips: physical-contradiction reframing, Ideal Final Result, structure-vs-physics, and a dissolve-vs-relocate test on every resolution. Refuses to return 'find the optimal balance', to fake a dissolution for a real law, or to call a cost-relocation a dissolution.

6 skills

Build an evidence-graded Goldratt Current-Reality Tree for a hard, contested problem: fan 20-55 AI agents across the branch-space, grade every node by the kind of evidence behind it, refute the weak branches, and converge on the ONE system constraint plus the cheapest test that would prove it. A heavyweight, multi-agent power instrument (needs Claude Code's Workflow tool) — not a course-pedagogy critic. Gates on a depth + token-cost choice before launching. Pairs with constraint-finder, which returns the 5 Focusing Steps once the constraint is located.

6 skills