Best Minds
- 893 installs
- 104 repo stars
- Updated January 7, 2026
- ceeon/best-minds
best-minds is a thinking-methodology skill that simulates the single best expert mind on a problem for developers who want domain-grounded answers instead of generic LLM opinions.
About
best-minds is a ceeon/best-minds auxiliary skill applying Andrej Karpathy's simulator framing: ask who in the world best understands the problem and respond from that expert's published ideas, writings, and quotes. It contrasts with preset coach lists by selecting experts based on problem essence, using one mind when sufficient or multiple only when collision adds value. Developers reach for best-minds during architecture debates, technology choices, or deep domain questions where a named expert's documented stance beats a generic assistant summary. Trigger terms include best minds, 顶级专家, 最强大脑, and 世界级.
- Shifts from generic LLM responses to targeted expert simulation
- Problem decides number of minds — one expert is usually enough
- Always selects the single strongest authority per question
- Grounds every response in the expert’s real public statements and writings
- Includes direct quotes from the chosen expert whenever possible
Best Minds by the numbers
- 893 all-time installs (skills.sh)
- +7 installs in the week ending Aug 5, 2026 (Skillselion tracking)
- Ranked #543 of 3,282 Productivity & Planning skills by installs in the Skillselion catalog
- Security screen: LOW risk (skills.sh audit)
- Data as of Aug 5, 2026 (Skillselion catalog sync)
npx skills add https://github.com/ceeon/best-minds --skill best-mindsAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 893 |
|---|---|
| repo stars | ★ 104 |
| Security audit | 3 / 3 scanners passed |
| Last updated | January 7, 2026 |
| Repository | ceeon/best-minds ↗ |
Who is the best expert to simulate for this problem?
Get world-class expert perspectives by simulating the thinking of the single best mind on any given problem instead of generic AI answers.
Who is it for?
Developers facing ambiguous architecture or domain decisions who want answers anchored to a specific recognized expert's public track record.
Skip if: Tasks requiring live data lookups, code generation, or compliance-sign-off documentation without expert simulation value.
When should I use this skill?
A developer asks who best understands a problem, requests a top expert perspective, or uses triggers like best minds, 顶级专家, or 世界级.
What you get
Expert-grounded answers with cited public ideas, writings, and original quotes from the selected domain authority.
By the numbers
- Contrasts with ai-coaches which selects from 13 preset personas
- Uses 4 core principles: problem decides count, find the strongest expert, base on real public work, and quote original s
Files
<!-- input: 用户问题 output: 领域专家视角的回答 pos: 辅助 skill，思维方法论
架构守护者：一旦我被修改，请同步更新： 1. 本文件的头部注释 2. 所属文件夹的 README.md -->
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-coaches | best-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 |
# OS
.DS_Store
Thumbs.db
*.swp
*.swo
*~
# IDE
.vscode/
.idea/
*.code-workspace
# Node (if needed)
node_modules/
package-lock.json
yarn.lock
# Environment
.env
.env.local
# Temporary files
*.tmp
*.temp
.cache/
MIT License
Copyright (c) 2025 Chengfeng
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
Related skills
Grill MeHave an agent relentlessly pressure-test their plan or design until every decision branch is resolved and shared understanding is reached.691k196kGrill With DocsPressure-test architecture decisions against their live codebase and living documentation before writing production code.585k196k
Lark CalendarLet their coding agents read, create, and manage events inside Lark (Feishu) Calendar directly from natural language instructions.493kLark ImLet their coding agent send instant notifications, post updates, and receive commands directly inside Lark (Feishu) workspaces.493kLark ContactLet their AI coding agents read and write contacts, send messages, and manage user data inside Lark (Feishu).493kLark AttendanceAutomatically fetch and report daily attendance data from Lark (Feishu) into their own tools or agents.493kHow it compares
Use best-minds for one deeply matched domain expert simulation rather than rotating through a fixed shortlist of generic coaches.
FAQ
How is best-minds different from preset AI coaches?
best-minds searches worldwide for the expert who truly understands the problem based on its essence, while preset coach skills pick from fixed lists such as 13 predefined personas matched by keywords rather than problem depth.
Does best-minds cite real expert sources?
best-minds grounds simulations in the selected expert's public thoughts, published works, and direct quotes, following principles to use original statements instead of inventing unattributed opinions.
Is Best Minds safe to install?
skills.sh reports 3 of 3 security scanners passed. Review the Security Audits panel on this page before installing in production.