Principle Redesign From First Principles
cursor/plugins
Apply when integrating a new requirement into an existing design.
Apply the Karpathy understanding-first principle — verify what AI built, own your understanding, map the jagged capability landscape of LLMs.
MITAuto-check passed
Install Karpathy Understanding First
$ npx skills add LearnPrompt/andrej-karpathy-skills --skill karpathy-understanding-first -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install LearnPrompt/andrej-karpathy-skills karpathy-understanding-first --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/LearnPrompt/andrej-karpathy-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/karpathy-understanding-first .claude/skills/karpathy-understanding-first && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "karpathy-understanding-first" agent skill from https://github.com/LearnPrompt/andrej-karpathy-skills/tree/main/karpathy-understanding-first into .claude/skills/karpathy-understanding-first/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "karpathy-understanding-first", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/LearnPrompt/andrej-karpathy-skills/tree/main/karpathy-understanding-firstType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add LearnPrompt/andrej-karpathy-skills --skill karpathy-understanding-first -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install LearnPrompt/andrej-karpathy-skills karpathy-understanding-first --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/LearnPrompt/andrej-karpathy-skills.git skills-src && mkdir -p .agents/skills && cp -r skills-src/karpathy-understanding-first .agents/skills/karpathy-understanding-first && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "karpathy-understanding-first" agent skill from https://github.com/LearnPrompt/andrej-karpathy-skills/tree/main/karpathy-understanding-first into .agents/skills/karpathy-understanding-first/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "karpathy-understanding-first", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add LearnPrompt/andrej-karpathy-skills --skill karpathy-understanding-first -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install LearnPrompt/andrej-karpathy-skills karpathy-understanding-first --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/LearnPrompt/andrej-karpathy-skills.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/karpathy-understanding-first .cursor/skills/karpathy-understanding-first && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "karpathy-understanding-first" agent skill from https://github.com/LearnPrompt/andrej-karpathy-skills/tree/main/karpathy-understanding-first into .cursor/skills/karpathy-understanding-first/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "karpathy-understanding-first", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/LearnPrompt/andrej-karpathy-skills.git --path karpathy-understanding-first--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add LearnPrompt/andrej-karpathy-skills --skill karpathy-understanding-first -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install LearnPrompt/andrej-karpathy-skills karpathy-understanding-first --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/LearnPrompt/andrej-karpathy-skills.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/karpathy-understanding-first .gemini/skills/karpathy-understanding-first && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "karpathy-understanding-first" agent skill from https://github.com/LearnPrompt/andrej-karpathy-skills/tree/main/karpathy-understanding-first into .gemini/skills/karpathy-understanding-first/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "karpathy-understanding-first", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install LearnPrompt/andrej-karpathy-skills karpathy-understanding-firstInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add LearnPrompt/andrej-karpathy-skills --skill karpathy-understanding-first -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/LearnPrompt/andrej-karpathy-skills.git skills-src && mkdir -p .github/skills && cp -r skills-src/karpathy-understanding-first .github/skills/karpathy-understanding-first && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "karpathy-understanding-first" agent skill from https://github.com/LearnPrompt/andrej-karpathy-skills/tree/main/karpathy-understanding-first into .github/skills/karpathy-understanding-first/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "karpathy-understanding-first", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add LearnPrompt/andrej-karpathy-skills --skill karpathy-understanding-first -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install LearnPrompt/andrej-karpathy-skills karpathy-understanding-first --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/LearnPrompt/andrej-karpathy-skills.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/karpathy-understanding-first .opencode/skills/karpathy-understanding-first && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "karpathy-understanding-first" agent skill from https://github.com/LearnPrompt/andrej-karpathy-skills/tree/main/karpathy-understanding-first into .opencode/skills/karpathy-understanding-first/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "karpathy-understanding-first", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
karpathy-understanding-first- GitHub stars
- 110
- Token cost
- ~1.6k tokens
- SKILL.md length
- 292 words
- Files
- 1
- Skills in repo
- 15
- Repo updated
- First seen
- Licence
- MIT
At a glance
Apply the Karpathy understanding-first principle — verify what AI built, own your understanding, map the jagged capability landscape of LLMs.
- The user has used AI to build something and needs to verify it
- SKILL.md covers Core Principle, The Understanding Audit, Jagged Capabilities Map and The Capability Check Prompt, plus 6 more sections
- Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
- Wants to audit their own understanding
What it does
Karpathy Understanding First is an agent skill from LearnPrompt/andrej-karpathy-skills. Apply the Karpathy understanding-first principle — verify what AI built, own your understanding, map the jagged capability landscape of LLMs. Use this skill when the user has used AI to build something and needs to verify it, wants to audit their own understanding, needs to know where LLMs are reliable vs unreliable, or says "verify this", "do I actually understand this", "what can I trust the AI on", "jagged capabilities", "understanding not outsourcing". Based on 46k-like quote post.
Its SKILL.md is about 1.6k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
The repository describes itself as: Karpathy-inspired Agent Skills collection. The licence is MIT.
When your agent uses it
- The user has used AI to build something and needs to verify it
- Wants to audit their own understanding
- Needs to know where LLMs are reliable vs unreliable
- Says verify this
Example prompts
- “verify this”
- “do I actually understand this”
- “what can I trust the AI on”
- “/karpathy-understanding-first”
What it can do on your machine
Read from SKILL.md and the folder at commit 9e46dec. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
No scripts in the folder and no shell commands in SKILL.md (its code samples are markdown).
From the folder's file list and the shell code blocks in SKILL.md.
Network
Links to these hosts (documentation or services it may open):
x.com
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Karpathy Understanding First loads about 1.6k tokens when it runs. Until then it costs about 130 tokens; SKILL.md has 292 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
The full file from LearnPrompt/andrej-karpathy-skills at commit 9e46dec, republished under its MIT licence (© LearnPrompt). 292 words, ~1,587 tokens.
Download SKILL.mdSave it as
.claude/skills/karpathy-understanding-first/SKILL.md (or your agent's skills folder).- name
- karpathy-understanding-first
- description
- Apply the Karpathy understanding-first principle — verify what AI built, own your understanding, map the jagged capability landscape of LLMs. Use this skill when the user has used AI to build something and needs to verify it, wants to audit their own understanding, needs to know where LLMs are reliable vs unreliable, or says "verify this", "do I actually understand this", "what can I trust the AI on", "jagged capabilities", "understanding not outsourcing". Based on 46k-like quote post.
- disable-model-invocation
- false
- user-invocable
- true
- related_skills
- karpathy-llm-simulator, karpathy-system-prompt-learning, karpathy-meta-reflection, karpathy-practice-environments
Skill 8: Understanding > Outsourcing（理解 > 外包）
Source: https://x.com/karpathy/status/2049907410303865030 "You can outsource your thinking but you cannot outsource your understanding." — 46k likes Also: Jagged Capabilities Modeling (Sequoia conversation)
Core Principle
Outsourcing is fine. Atrophy is not.
Use AI to go 10x faster. But every time you skip understanding something, you're borrowing from your own future. Understanding compounds. Blindly shipping agent output does not.
Karpathy's "jagged capabilities" insight: LLMs are wildly good at some things (verifiable, high-data domains) and silently wrong at others (off-distribution, novel combinations). You need a mental map of which is which.
The Understanding Audit
After every agent-assisted work session, run this:
I just had an AI help me [DESCRIBE WHAT WAS BUILT/WRITTEN/DECIDED].
Help me audit my own understanding:
1. List the 5 key decisions that were made (technical choices, assumptions, trade-offs)
2. For each decision: would I be able to explain WHY this choice was made to a colleague?
Mark each: UNDERSTOOD / SHALLOW / BLACK BOX
3. For the BLACK BOX items: give me a 3-sentence explanation I can verify
4. What would I need to read/learn to make the SHALLOW items UNDERSTOOD?
Be honest. If I'm shipping something I don't understand, I need to know.Jagged Capabilities Map
LLMs are NOT uniformly capable. Use this mental model:
HIGH Reliability (on the rails)
These are well-defined, verifiable, data-rich domains:
- Code in popular languages (Python, JS, SQL)
- Text summarization and rewriting
- Structured data extraction
- Well-known algorithms
- Common patterns (REST APIs, SQL queries, regex)
MEDIUM Reliability (use carefully, verify)
- Less popular languages (Rust, Haskell, Elixir)
- Integrating multiple systems together
- Anything involving specific versions/APIs after training cutoff
- Multi-step reasoning chains
- Math beyond simple arithmetic
LOW Reliability (always verify independently)
- Cutting-edge research (may hallucinate papers)
- Novel combinations of technologies
- Domain-specific knowledge you can't easily verify
- Legal/medical/financial specifics
- Anything where being wrong is expensive
The Capability Check Prompt
Before trusting an AI output on something important:
I need to evaluate how reliable your output is for this specific task:
Task type: [DESCRIBE WHAT YOU ASKED THE AI]
Output: [PASTE THE OUTPUT]
Help me assess:
1. How common is this exact pattern in training data? (ubiquitous / common / rare / novel)
2. What's the easiest way to verify this without running it?
3. What's the most likely error — hallucinated API? Wrong logic? Outdated syntax?
4. What search query would I use to double-check the key claim?
5. Should I trust this output? (YES/VERIFY_FIRST/DEFINITELY_CHECK)The Anti-Atrophy Protocol
Karpathy tracks which skills are atrophying (taken over by AI) and which are growing. Monthly check:
Run a skills audit for [MONTH/YEAR]:
Things I've been having AI do for me recently:
[LIST 5-10 TASKS]
For each:
1. Is my ability to do this WITHOUT AI declining? (YES / NO / UNSURE)
2. Is that OK? (Critical skill I must maintain / Commodity I'm fine outsourcing)
3. Action: KEEP_DELEGATING / PRACTICE_MANUALLY / LEARN_DEEPLY
Final question: What's one thing I should deliberately do WITHOUT AI this month to stay sharp?The Verification Checklist
Before shipping any agent-produced output:
Understanding verification:
- [ ] I can explain what this does in plain English without looking at it
- [ ] I've traced through the logic manually for at least one example
- [ ] I know what the failure mode looks like
- [ ] I've run it / tested it myself (not just read the output)
- [ ] If it's wrong, I'll know how to debug it
- [ ] I'm not shipping something I'd be embarrassed to explain in a code review
If any box is unchecked: understand it before shipping.The Learning Extraction Prompt
When AI writes something you don't understand — learn from it, don't just ship it:
You wrote this code/text and I don't fully understand it:
[PASTE OUTPUT]
Teach me:
1. The underlying concept (assume I know [MY_LEVEL] — beginner/intermediate/expert)
2. Walk through the logic step by step
3. What would happen if I changed [SPECIFIC PART]?
4. Point me to the canonical resource where I can learn this properly
5. Give me a simpler version that demonstrates the core idea in isolation
I'm going to use this as a learning moment, not just copy-paste.Workflow
属于工作流：反偏见决策（第2步）+ 工作流：月度体检（第2步）
| 位置 | 上游 | 下游 |
|---|---|---|
| C第2步 | karpathy-llm-simulator（辩论后） | karpathy-system-prompt-learning（沉淀规则） |
| D第2步 | karpathy-meta-reflection（审计后） | karpathy-practice-environments（建练习环境） |
反偏见决策链路：llm-simulator → understanding-first → system-prompt-learning 月度体检链路：meta-reflection → understanding-first → practice-environments → education-first
Prompt Contract
After completing this task, append an Understanding Report: 1) Key assumptions made (list each), 2) What was verified (with evidence), 3) What remains inferred/unverified (with suggested verification method), 4) Capability boundary notes (where LLM might be off-distribution for this task), 5) Items I must personally inspect or understand before relying on this output.Verification Checklist
- 关键假设被明确列出（不藏在输出里）
- 已验证项有证据（测试通过/文档引用/数据对比）
- 未验证项有明确的验证方法建议
- LLM 能力边界有标注（哪些部分它可能不擅长）
- 「你必须亲自看」清单具体到文件/行/配置
- 人类确认已理解核心逻辑（不只是看了一眼）
© LearnPrompt, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
Just SKILL.md in karpathy-understanding-first of LearnPrompt/andrej-karpathy-skills.
Open the folder on GitHubat commit 9e46dec
Compare with similar skills
Karpathy Understanding First next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Karpathy Understanding First this skillLearnPrompt/andrej-karpathy-skills | 110 | — | ~1.6k | Automated safety check: Pass | MIT | |
| Principle Redesign From First Principlescursor/plugins | 10k | 8 repos | ~211 | Automated safety check: Pass | None | |
| Video Understandcalesthio/OpenMontage | 66k | — | ~841 | Automated safety check: Pass | AGPL-3.0 | |
| Uxui Principlessickn33/agentic-awesome-skills | 47k | 2 repos | ~548 | Automated safety check: Pass | MIT | |
| Understand ExplainEgonex-AI/Understand-Anything | 86k | — | ~1.3k | Automated safety check: Pass | MIT | |
| Karpathy Principlesathola/claude-night-market | 341 | — | ~2k | Automated safety check: Pass | MIT |
Similar skills
- Official
Video Understand
calesthio/OpenMontage
Understand video content locally using ffmpeg frame extraction and Whisper transcription.
Uxui Principles
sickn33/agentic-awesome-skills
Evaluate interfaces against 168 research-backed UX/UI principles, detect antipatterns, and inject UX context into AI coding sessions.
Understand Explain
Egonex-AI/Understand-Anything
Gives an in-depth explanation of one file, function or module by reading the project's knowledge graph and checking that the graph is still fresh.
Karpathy Principles
athola/claude-night-market
Pre-implementation gate covering think-first, simplicity, surgical edits, and verifiable goals.
Understanding by Design Planner
THU-MAIC/OpenMAIC
Plans an OpenMAIC lesson or course series by backward design: enduring understandings and essential questions first, then performance evidence, then learning activities.
More from LearnPrompt/andrej-karpathy-skills
All 15 skills in this repoKarpathy Methodology Index
LearnPrompt/andrej-karpathy-skills
Apply Andrej Karpathy AI methodology and principles from his 2023-2026 insights. Use this skill when the user wants to apply Karpathy-style thinking, needs…
Karpathy Agentic Engineering
LearnPrompt/andrej-karpathy-skills
Apply Karpathy-style agentic engineering to any coding or building task.
AutoResearch Loop
LearnPrompt/andrej-karpathy-skills
Sets up an autonomous research loop where an agent runs experiments on git branches, logs results and proposes the next iteration while you approve each hypothesis change.
Karpathy Education First
LearnPrompt/andrej-karpathy-skills
Apply the education-first mindset — make everything you build teachable, create nano-project explanations, write for beginners.
Karpathy Idea Files
LearnPrompt/andrej-karpathy-skills
Create and share ideas as abstract Gist-style specs instead of code — letting agents or others implement.
Karpathy LLM Simulator
LearnPrompt/andrej-karpathy-skills
Use LLM as a simulator of expert debates and opposing viewpoints instead of getting a single sycophantic answer.
Questions about Karpathy Understanding First
What does Karpathy Understanding First do?
Apply the Karpathy understanding-first principle — verify what AI built, own your understanding, map the jagged capability landscape of LLMs. Karpathy Understanding First is an agent skill from LearnPrompt/andrej-karpathy-skills. Apply the Karpathy understanding-first principle — verify what AI built, own your understanding, map the jagged capability landscape of LLMs.
When should I use Karpathy Understanding First?
Karpathy Understanding First fits situations like: the user has used AI to build something and needs to verify it; wants to audit their own understanding; needs to know where LLMs are reliable vs unreliable; says verify this.
How do I install Karpathy Understanding First in Claude Code?
Run `npx skills add LearnPrompt/andrej-karpathy-skills --skill karpathy-understanding-first -a claude-code`. Or copy the skill folder (karpathy-understanding-first in LearnPrompt/andrej-karpathy-skills) into .claude/skills/karpathy-understanding-first in your project. Claude Code loads it when a task matches its description.
How do I install Karpathy Understanding First in Codex?
Run `npx skills add LearnPrompt/andrej-karpathy-skills --skill karpathy-understanding-first -a codex`. Or copy the skill folder (karpathy-understanding-first in LearnPrompt/andrej-karpathy-skills) into .agents/skills/karpathy-understanding-first in your project. Codex loads it when a task matches its description.
Can I use Karpathy Understanding First in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add LearnPrompt/andrej-karpathy-skills --skill karpathy-understanding-first -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/karpathy-understanding-first, .gemini/skills/karpathy-understanding-first, .github/skills/karpathy-understanding-first and .opencode/skills/karpathy-understanding-first in your project.
What does Karpathy Understanding First need to run?
SKILL.md names no scripts, command-line tools or credentials: Karpathy Understanding First is instructions for the agent only.
Does Karpathy Understanding First access the network?
SKILL.md names 1 domain. As links in the text: x.com. This is read from the text; nothing was executed.
Is Karpathy Understanding First safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Karpathy Understanding First use?
Karpathy Understanding First is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Karpathy Understanding First use?
About 1.6k tokens (SKILL.md is roughly 6.3k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Karpathy Understanding First?
Skills that share tags, products or a category with Karpathy Understanding First: Principle Redesign From First Principles (cursor/plugins, 10k stars), Video Understand (calesthio/OpenMontage, 66k stars), Uxui Principles (sickn33/agentic-awesome-skills, 47k stars) and Understand Explain (Egonex-AI/Understand-Anything, 86k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Karpathy Understanding First?
LearnPrompt (a GitHub user) maintains it in LearnPrompt/andrej-karpathy-skills, which has 110 GitHub stars. The repository holds 15 skills in this directory. The repository was last updated on July 10, 2026.
Source: LearnPrompt/andrej-karpathy-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.