28 Mental Models, Zero Proven to Help

Added 2026-08-06

The entry point is a routing meta-skill, with another for combining models. The theory is credited by name to Munger, Meadows, Kahneman, Goldratt, TRIZ and OODA.

EngineeringContent & KnowledgeProduct & Project

View on GitHub ↗

28 Mental Models, Zero Proven to Help preview

The author leads with the negative result: no skill holds a robust, replicated verdict of improvement, and all twenty-eight are manual-only. The closest one ends at directional-but-not-replicated. Thirty-nine were cut to twenty-eight after evidence review. His own conclusion: treat them as scaffolding for structured reasoning, not a guarantee of accuracy. The evaluation apparatus is real: structural validation, routing false-trigger control, blind pairwise evals, and a provenance register that fails closed.

Related skills

Devil's Advocate: Only Challenge, No ComfortThe whole repository is three files, and the core is a rule set that inverts the model's default agreeableness. The claim is blunt: the danger is not that AI errs, but that it ships a bad idea with equal enthusiasm.Content & Knowledge · Product & Project +128 stars Lark's Official CLI, in Three Command TiersPlus-prefixed shortcuts serve humans and agents alike, with smart defaults and dry-run; 100-plus API commands map one-to-one onto platform endpoints; below that, lark-cli api reaches 2,500-plus endpoints raw.Product & Project · Content & Knowledge +116.2K stars Six Skills Khazix Uses HimselfSix skills, one job each: break down a goal, clean up disk space, write research reports, track what is trending, tidy files, and write. From Khazix, founder of Xushi Media.Content & Knowledge · Product & Project19.3K stars Small, Hackable Skills for Real EngineeringThe skills Matt Pocock uses daily: get grilled until nothing is ambiguous, turn talk into a spec and linked tickets, build red-green-refactor, run hard bugs through a fixed loop. Small and hackable, not process-owning.Engineering · Product & Project205.2K stars