cc-thinking-skills
Structured reasoning frameworks for AI agents to decide and diagnose
All Skills in This Repository (39)
Pure Emerald Level Indicators
tjboudreauxthinking-socratic
Systematically applies six Socratic question types to deepen understanding and surface hidden assumptions in analysis and decision-making.
tjboudreauxthinking-fermi-estimation
Decompose unknown quantities into estimable factors and combine them for order-of-magnitude estimates.
tjboudreauxthinking-effectuation
Coordinate uncertain startup ventures using effectuation principles.
tjboudreauxthinking-debiasing
Apply a 12-point checklist to identify and mitigate cognitive biases in decisions.
tjboudreauxthinking-jobs-to-be-done
Identify job performers, map job steps, and define outcome metrics.
tjboudreauxthinking-leverage-points
Map interventions using Meadows' 12 leverage points to prioritize strategic changes.
tjboudreauxthinking-steel-manning
Construct and defend the strongest opposing argument to validate decisions.
tjboudreauxthinking-bayesian
Apply Bayes' theorem to update beliefs with explicit priors and likelihoods.
tjboudreauxthinking-ooda
Guide rapid decision-making through Observe, Orient, Decide, and Act phases.
tjboudreauxthinking-theory-of-constraints
Identify system bottlenecks and apply the five focusing steps to improve throughput.
tjboudreauxthinking-triz
Resolve technical contradictions using TRIZ principles and the 40 inventive principles.
tjboudreauxthinking-inversion
Convert failure modes into explicit requirements and testable guardrails.
Frequently Asked Questions
FAQPage SchemaHow to install cc-thinking-skills?▼
Run `npx skills add tjboudreaux/cc-thinking-skills --all -g -y` in your terminal to install all 28 thinking skills globally.
What do thinking skills do for AI agents?▼
They give your agent step-by-step reasoning procedures for tasks like debugging root causes, weighing decisions, and stress-testing plans, instead of relying on unstructured guessing.
Which thinking skill should I start with?▼
Start with thinking-model-router when you are unsure which framework fits; it picks the right skill or returns none. Invoke a specific skill directly when the match is obvious.
Do these skills work with Claude Code and Cursor?▼
Yes. All skills follow the portable SKILL.md standard and load in Claude Code, GitHub Copilot, Codex, Cursor, and other compatible tools.
Do thinking skills guarantee better AI accuracy?▼
No. The project's own evidence audit supports structured procedures but does not claim a guaranteed accuracy gain, and every skill is validated with evaluation runs.
