Sandbox
@mattnowdev/thinking-partner

Thinking partner skill for agent decision making

This skill adds a structured thinking partner to your agent conversations. It challenges assumptions, detects orientation capture, and applies the right mental models instead of offering generic advice. The workflow is deterministic, with a catalog of 150+ models organized by discipline.

190 stars14 forksUpdated 5mo ago
Who it's for

Builders who want Claude Code, Cursor, Windsurf, or Copilot to help them think through hard decisions.

What it delivers

You can stress test ideas, compare options, and catch hidden assumptions before you act.

What it does

Assumption checking

Surfaces hidden beliefs you may be treating as facts.

Mental model selection

Chooses from 150+ models across 17 disciplines, including first principles, inversion, pre-mortems, and 5 Whys.

Orientation detection

Detects six thinking states and adjusts the intervention to match.

Cognitive operations

Uses paired operations like Decouple/Recouple and Hold/Resolve to keep thinking productive.

Deterministic workflow

Follows a fixed six-step process from diagnosis to synthesis instead of freeform questioning.

Anti-pattern guidance

Includes explicit guidance on what not to do when challenging a problem.

How to get it

  1. 1Run
    npx skills add mattnowdev/thinking-partner
  2. 2Claude Code
    git clone https://github.com/mattnowdev/thinking-partner.git
    cp -r thinking-partner/skills/thinking-partner ~/.claude/skills/thinking-partner
  3. 3Cursor
    git clone https://github.com/mattnowdev/thinking-partner.git
    cp -r thinking-partner/skills/thinking-partner .cursor/skills/thinking-partner
  4. 4Windsurf
    git clone https://github.com/mattnowdev/thinking-partner.git
    cp -r thinking-partner/skills/thinking-partner .windsurf/skills/thinking-partner

README

Thinking Partner

A deterministic thinking partner that challenges assumptions and applies 150+ mental models to sharpen decisions, solve problems, and think more clearly.

Built on the open Agent Skills standard. Works with Claude Code, Cursor, Windsurf, Cline, GitHub Copilot, and any AI agent that supports the standard.

Install

npx skills add mattnowdev/thinking-partner

What It Does

Not a lecture — a sparring session. Turns your AI agent into a thinking partner that:

  • Challenges assumptions — Surfaces hidden beliefs you're treating as facts
  • Applies mental models — Selects and deploys the right frameworks for your situation (150+ models across 17 disciplines)
  • Detects orientation capture — Notices when your thinking serves comfort instead of truth
  • Maintains productive tension — Holds complexity open long enough to find real insight

When It Triggers

  • "Help me think through X"
  • "Challenge my thinking"
  • "What am I missing?"
  • "Play devil's advocate"
  • "Stress test this idea"
  • "Help me decide between X and Y"
  • Any named mental model: SWOT, first principles, inversion, pre-mortem, 5 Whys, etc.

What Makes It Different

FeatureThis SkillOthers
Orientation DetectionDiagnoses 6 thinking states (GT0-GT5) with targeted interventionsGeneric questioning
150+ Mental ModelsFull catalog organized by discipline with key questionsHandful of models
Cognitive Operations7 complementary operation pairs (Decouple/Re-couple, Hold/Resolve, etc.)None
Structured Workflow6-step deterministic process from diagnosis to synthesisFreeform
Anti-PatternsExplicit guidance on what NOT to doNone

Mental Models Included

Organized across 17 disciplines:

  • General Thinking: First Principles, Inversion, Second-Order Thinking, Chesterton's Fence, Entropy, and more
  • Decision-Making: Pre-Mortem, Regret Minimization, SWOT, Asymmetric Risk, Preserving Optionality
  • Problem-Solving: 5 Whys, Fishbone, MECE Decomposition, Bright Spots Analysis, Local vs Global Optima
  • Systems & Complexity: Feedback Loops, Stocks & Flows, Tipping Points, Antifragility, Path Dependence
  • Economics & Strategy: Incentives, Network Effects, Tragedy of the Commons, Lindy Effect, Power Laws
  • Statistics & Probability: Bayesian Updating, Correlation vs Causation, Selection Bias, Gambler's Fallacy
  • Psychology & Biases: Confirmation Bias, Planning Fallacy, Halo Effect, Curse of Knowledge, Peak-End Rule
  • Physics & Engineering: Inertia, Activation Energy, Resonance, Half-Life, Margin of Safety
  • Biology & Evolution: Natural Selection, Red Queen Effect, Coevolution, Niche Construction
  • Communication & Persuasion: Steel Manning, Reciprocity, Narrative/Storytelling, Pyramid Principle
  • Creativity & Innovation: SCAMPER, Design Thinking, Oblique Strategies, Minimum Viable Experiment
  • Learning & Development: Feynman Technique, Spaced Repetition, Zone of Proximal Development
  • Time & Resource Management: Eisenhower Matrix, Pareto Principle, Maker's vs Manager's Schedule
  • Game Theory: Prisoner's Dilemma, Schelling Point, Tit for Tat, Signaling, Moral Hazard
  • Negotiation: BATNA, ZOPA, Logrolling
  • Resilience & Antifragility: Barbell Strategy, Skin in the Game, Via Negativa, Hormesis
  • Ethics & Responsibility: Veil of Ignorance

See the full catalog in skills/thinking-partner/references/model-catalog.md.

Alternative Install

Manual install per agent

Claude Code

git clone https://github.com/mattnowdev/thinking-partner.git
cp -r thinking-partner/skills/thinking-partner ~/.claude/skills/thinking-partner

Cursor

git clone https://github.com/mattnowdev/thinking-partner.git
cp -r thinking-partner/skills/thinking-partner .cursor/skills/thinking-partner

Windsurf

git clone https://github.com/mattnowdev/thinking-partner.git
cp -r thinking-partner/skills/thinking-partner .windsurf/skills/thinking-partner

Other Agents Copy the skills/thinking-partner/ directory (including references/) into your agent's skills directory.

License

MIT

Files in the repo

Repository payload5 top-level entries
  • .claude-plugin
  • skills
  • .gitignore
  • LICENSE
  • README.md

Discussion (0)

Ask about usage, or say what you built with it

Sign in to join the discussion.

No comments yet. Be the first to say what this is good for.

More skills

Turn any codebase, with its docs, SQL schemas, configs, and PDFs, into a queryable knowledge graph. A /graphify skill for Claude Code, Cursor, Codex, and Gemini CLI: local deterministic AST parsing, every edge explained, no vector store.

117k
1 add

Open-source AI job search: scan job portals, evaluate listings into a structured A-H report with a global 1-5 score, tailor your CV, track applications — runs locally in your AI coding CLI (Claude Code, Codex, OpenCode, Antigravity…)

71k
mvanhorn/
last30days-skill

AI agent skill that researches any topic across Reddit, X, YouTube, HN, Polymarket, and the web - then synthesizes a grounded summary

62k
Imbad0202/
academic-research-skills

Academic Research Skills for Claude Code: research → write → review → revise → finalize

48k

AI video skill for Claude Code & Codex — cinematic product videos with Remotion: 152 shot recipe cards, 209 motion previews, a production-ready template

8k

817 structured cybersecurity skills for AI agents · Mapped to 6 frameworks: MITRE ATT&CK, NIST CSF 2.0, MITRE ATLAS, D3FEND, NIST AI RMF & MITRE F3 (Fight Fraud) · agentskills.io standard · Works with Claude Code, GitHub Copilot, Codex CLI, Cursor, Gemini CLI & 20+ platforms · 29 security domains · Apache 2.0

33k