Claude MarketplacesSubmit
brucexo avatar

Best Minds

brucexo/skills-collection

Instead of asking Claude "what do you think," this prompts you to ask "who knows this best in the world, and what would they say?" It's based on Andrej Karpathy's idea of treating LLMs as simulators rather than entities. You specify the domain and it identifies the top expert, then simulates their perspective using their actual published work and statements. Works with trigger phrases like "best minds" or "who knows this best." The approach is simple: find the strongest voice on a topic, not just a suitable one, and ground the simulation in real quotes and ideas. Use one expert when that's enough, multiple when you need perspectives to collide.

skills.sh installs
246
GitHub stars
2
last updated
Jun 3, 2026
View on GitHub

Install

$ npx -y skills add brucexo/skills-collection --skill best-minds --agent claude-code

Installs into .claude/skills of the current project.

Files

Best Minds

"Don't think of LLMs as entities but as simulators." — Andrej Karpathy

核心

不要问 AI "你怎么看"。

要问：这个问题，世界上谁最懂？TA 会怎么说？

然后模拟那个人。

原则

  1. 问题决定人数 — 一个人够就一个，需要碰撞才多个
  2. 找真正最懂的 — 不是找"合适的"，是找"最强的"
  3. 基于真实 — 模拟要基于 TA 公开的思想、著作、言论
  4. 引用原话 — 尽可能用 TA 说过的话

vs ai-coaches

ai-coachesbest-minds
从 13 个预设智者选从全世界找
基于关键词匹配基于问题本质
SKILL.md on GitHub

Sponsors

Promote your product hereReach 125k+ visitors monthlyPromote your product hereReach 125k+ visitors monthlyAgent CaffeineSponsoredClose your lid, come back to finished workSnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthlyWaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthlyknobsSponsoredSee your site on any phone, any day, in any language

Details

Repository
github.com/brucexo/skills-collection
First seen
Jun 3, 2026

Recommended

1juliusbrussee avatarcavemanjuliusbrussee/cavemanUltra-compressed communication mode. Cuts output tokens 65% (measured) by speaking like caveman while keeping full technical accuracy. Supports intensity levels618.3Kskills.sh installs
2mattpocock avatargrill-memattpocock/skillsA relentless interview to sharpen a plan or design.1.5Mskills.sh installs
3shadcn avatarimproveshadcn/improveSurvey any codebase as a senior advisor and produce prioritized, self-contained implementation plans for OTHER models/agents to execute. Strictly read-only on s40Kskills.sh installs
4obra avatarsystematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixes326.7Kskills.sh installs
5forrestchang avatarkarpathy-guidelinesforrestchang/andrej-karpathy-skillsBehavioral guidelines to reduce common LLM coding mistakes through explicit assumptions, simplicity, and verifiable success criteria.17.9Kskills.sh installs
6vercel-labs avatarfind-skillsvercel-labs/skillsHelps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express int4.3Mskills.sh installs