Karpathy Guidelines
alirezarezvani/ClaudeForge
Behavioral guardrails for LLM-assisted coding. An agent skill from alirezarezvani/ClaudeForge.
This skill channels the strategic thinking of Mark Zuckerberg, co-founder and CEO of Meta.
Install Mark Zuckerberg
$ npx skills add K-Dense-AI/mimeographs --skill mark-zuckerberg -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install K-Dense-AI/mimeographs mark-zuckerberg --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/K-Dense-AI/mimeographs.git skills-src && mkdir -p .claude/skills && cp -r skills-src/mimeographs/mark-zuckerberg .claude/skills/mark-zuckerberg && 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 "mark-zuckerberg" agent skill from https://github.com/K-Dense-AI/mimeographs/tree/main/mimeographs/mark-zuckerberg into .claude/skills/mark-zuckerberg/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "mark-zuckerberg", 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/K-Dense-AI/mimeographs/tree/main/mimeographs/mark-zuckerbergType 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 K-Dense-AI/mimeographs --skill mark-zuckerberg -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install K-Dense-AI/mimeographs mark-zuckerberg --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/K-Dense-AI/mimeographs.git skills-src && mkdir -p .agents/skills && cp -r skills-src/mimeographs/mark-zuckerberg .agents/skills/mark-zuckerberg && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "mark-zuckerberg" agent skill from https://github.com/K-Dense-AI/mimeographs/tree/main/mimeographs/mark-zuckerberg into .agents/skills/mark-zuckerberg/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "mark-zuckerberg", 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 K-Dense-AI/mimeographs --skill mark-zuckerberg -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install K-Dense-AI/mimeographs mark-zuckerberg --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/K-Dense-AI/mimeographs.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/mimeographs/mark-zuckerberg .cursor/skills/mark-zuckerberg && 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 "mark-zuckerberg" agent skill from https://github.com/K-Dense-AI/mimeographs/tree/main/mimeographs/mark-zuckerberg into .cursor/skills/mark-zuckerberg/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "mark-zuckerberg", 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/K-Dense-AI/mimeographs.git --path mimeographs/mark-zuckerberg--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 K-Dense-AI/mimeographs --skill mark-zuckerberg -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install K-Dense-AI/mimeographs mark-zuckerberg --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/K-Dense-AI/mimeographs.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/mimeographs/mark-zuckerberg .gemini/skills/mark-zuckerberg && 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 "mark-zuckerberg" agent skill from https://github.com/K-Dense-AI/mimeographs/tree/main/mimeographs/mark-zuckerberg into .gemini/skills/mark-zuckerberg/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "mark-zuckerberg", 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 K-Dense-AI/mimeographs mark-zuckerbergInstalls 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 K-Dense-AI/mimeographs --skill mark-zuckerberg -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/K-Dense-AI/mimeographs.git skills-src && mkdir -p .github/skills && cp -r skills-src/mimeographs/mark-zuckerberg .github/skills/mark-zuckerberg && 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 "mark-zuckerberg" agent skill from https://github.com/K-Dense-AI/mimeographs/tree/main/mimeographs/mark-zuckerberg into .github/skills/mark-zuckerberg/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "mark-zuckerberg", 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 K-Dense-AI/mimeographs --skill mark-zuckerberg -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install K-Dense-AI/mimeographs mark-zuckerberg --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/K-Dense-AI/mimeographs.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/mimeographs/mark-zuckerberg .opencode/skills/mark-zuckerberg && 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 "mark-zuckerberg" agent skill from https://github.com/K-Dense-AI/mimeographs/tree/main/mimeographs/mark-zuckerberg into .opencode/skills/mark-zuckerberg/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "mark-zuckerberg", 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
mark-zuckerberg- GitHub stars
- 129
- Token cost
- ~1.4k tokens
- SKILL.md length
- 696 words
- Files
- 72 (incl. references)
- Skills in repo
- 60
- Repo updated
- First seen
- Licence
- MIT
At a glance
This skill channels the strategic thinking of Mark Zuckerberg, co-founder and CEO of Meta.
- Works in 3 steps: Stop debating and build a prototype… → Release a small iteration quickly rather… → Foster an open meritocracy where the…
- Apply his frameworks on decentralized testing
- SKILL.md covers Core principles, How Mark Zuckerberg reasons, Applying the frameworks and Anti-patterns they push against, plus 1 more section
- Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
What it does
Mark Zuckerberg is an agent skill from K-Dense-AI/mimeographs. This skill channels the strategic thinking of Mark Zuckerberg, co-founder and CEO of Meta. Reach for this skill whenever the user is navigating rapid product iteration, scaling social platforms, open-source AI strategy, content moderation trade-offs, or long-term organizational management. It is highly effective for scenarios involving "The Hacker Way" (rapid prototyping), building community, handling intense public criticism, or deciding between open vs. closed ecosystems. Use this skill to apply his frameworks…
Its SKILL.md is about 1.4k tokens, which your agent loads only when the skill is triggered. The skill folder holds 74 other files, including reference files (for example `AGENTS.md`, `_workspace/agents_output.e584bd6c.json` and `_workspace/clustered_corpus.e584bd6c.json`).
It sits in Development, covering Prototyping and LLM guardrails. The repository describes itself as: Ready-to-use agent skills that clone the thinking of founders, philosophers, and scientists into your agent. Generated with K-Dense-AI/mimeo. The licence is MIT.
When your agent uses it
- Apply his frameworks on decentralized testing
- Small group context sharing
- Progressive AI integration
- Even if the user doesnt explicitly mention his name
Example prompts
- “The Hacker Way”
- “/mark-zuckerberg”
Workflow steps
3 steps, taken from the first numbered list in SKILL.md.
- Stop debating and build a prototype immediately ("Code wins arguments").
- Release a small iteration quickly rather than trying to get everything right at once.
- Foster an open meritocracy where the best implementation wins, regardless of hierarchy.
What it can do on your machine
Read from SKILL.md and the folder at commit a38f5fc. 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.
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
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
Mark Zuckerberg loads about 1.4k tokens when it runs, and up to ~5.5k if it reads all its reference files. Until then it costs about 169 tokens; SKILL.md has 696 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 K-Dense-AI/mimeographs at commit a38f5fc, republished under its MIT licence (© K-Dense-AI). 696 words, ~1,379 tokens.
Download SKILL.mdSave it as
.claude/skills/mark-zuckerberg/SKILL.md (or your agent's skills folder). This skill also uses 71 other files; get the full folder from GitHub.- name
- mark-zuckerberg
- description
- This skill channels the strategic thinking of Mark Zuckerberg, co-founder and CEO of Meta. Reach for this skill whenever the user is navigating rapid product iteration, scaling social platforms, open-source AI strategy, content moderation trade-offs, or long-term organizational management. It is highly effective for scenarios involving "The Hacker Way" (rapid prototyping), building community, handling intense public criticism, or deciding between open vs. closed ecosystems. Use this skill to apply his frameworks on decentralized testing, small group context sharing, and progressive AI integration, even if the user doesn't explicitly mention his name.
Thinking like Mark Zuckerberg
Mark Zuckerberg's thinking is characterized by a relentless focus on long-term utility, rapid iteration, and massive scale. He views technology not as an end in itself, but as infrastructure for human connection and community building. His strategic approach often involves making massive, multi-year bets on future platforms (like the Metaverse or open-source AI) while relying on a highly decentralized, hacker-driven culture to figure out the immediate steps to get there.
Reach for this skill whenever you're advising on product-market fit at scale, navigating the trade-offs of open vs. closed systems, designing company culture and values, or managing an organization through intense public scrutiny.
Core principles
- Ideas Don't Come Out Fully Formed: Action brings clarity; waiting until you understand everything about a complex problem guarantees you will never start.
- Make Money to Build Better Services: The business engine exists to support the social mission and long-term investments, not the other way around.
- Values Must Be Debatable: Generic statements like "be honest" are useless; values must represent a specific cultural operating choice that someone could legitimately disagree with.
- The Alternate Universe Hiring Test: Only hire someone if you would be happy working for them in an alternate universe where the roles were reversed.
- Platforms Should Not Be Arbiters of Truth: Social media companies exist to give people a voice; deciding what is true and false for billions of people destroys public trust.
For detailed rationale and quotes, see references/principles.md.
How Mark Zuckerberg reasons
Zuckerberg reasons by looking at the longest possible time horizon and working backward to the immediate prototype. He prioritizes utility over "coolness" and believes that solving physical world problems is often best done by improving digital infrastructure (Bits vs. Atoms). When evaluating products, he focuses on "Time Well Spent vs. Time Spent," prioritizing meaningful interactions over passive consumption, even if it hurts short-term metrics.
During crises, his reasoning shifts into "Wartime CEO" mode, where he abandons consensus-building in favor of rapid, unilateral decisions to ensure survival. He manages his own attention carefully, aware of "The Eye of Sauron" effect, where intense leadership scrutiny can burn out a team. For a full catalog of his mental models, see references/mental-models.md.
Applying the frameworks
The Hacker Way (Continuous Iteration)
When to use: The user is stuck debating a product feature or trying to perfect a release.
- Stop debating and build a prototype immediately ("Code wins arguments").
- Release a small iteration quickly rather than trying to get everything right at once.
- Foster an open meritocracy where the best implementation wins, regardless of hierarchy.
Show full SKILL.md (269 more words)Show less
Small Group Context Sharing
When to use: The user is struggling with execution and agility in a growing organization.
- Identify a core group of 25-30 key leaders across the company.
- Spend significant time sharing deep context with this group rather than relying on rigid hierarchical delegation.
- Deploy this highly connected, high-trust group as a "core army" to execute cross-functional priorities.
For more frameworks, including Precision vs. Recall Moderation Tuning and Progressive Model Integration, see references/frameworks.md.
Anti-patterns they push against
- Waiting for the "Eureka Moment": Believing ideas come fully formed prevents you from ever getting started.
- Creating Generic Company Values: If no one can legitimately disagree with a value, it cannot guide hard decisions.
- Slowing Down Out of Fear of Mistakes: If you never break anything, you aren't taking enough risks to reach your potential.
- Consensus-Building During a Crisis: Prevents the rapid, unilateral decision-making required to survive existential threats.
- Keeping AI Models Closed: Creates a massive vulnerability where a single bad actor with superior AI can compromise global systems.
How to use this skill in conversation
When the user is facing a situation involving scale, iteration, or platform strategy, surface the relevant principle or framework by name. Apply it directly to their context and cite where the idea comes from (e.g., "Mark Zuckerberg calls this the Alternate Universe Hiring Test, which means...").
Do not impersonate Zuckerberg or speak in the first person. Instead, channel his analytical, long-term, and utility-driven perspective. If the user is stuck in analysis paralysis, push them toward "The Hacker Way." If they are struggling with company culture, introduce the concept of "Debatable Values."
© K-Dense-AI, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
SKILL.md and 71 other files (references) in mimeographs/mark-zuckerberg of K-Dense-AI/mimeographs.
- SKILL.md
- AGENTS.md
- _workspace/agents_output.e584bd6c.json
- _workspace/clustered_corpus.e584bd6c.json
- _workspace/discovery/books.json
- _workspace/discovery/essays.json
- _workspace/discovery/frameworks.json
- _workspace/discovery/interviews.json
- _workspace/discovery/letters.json
- _workspace/discovery/papers.json
- _workspace/discovery/podcasts.json
- _workspace/discovery/ranked_sources.e584bd6c.json
- _workspace/discovery/talks.json
- _workspace/distilled/src_001.e584bd6c.json
- _workspace/distilled/src_014.e584bd6c.json
- _workspace/distilled/src_016.e584bd6c.json
- _workspace/distilled/src_017.e584bd6c.json
- _workspace/distilled/src_020.e584bd6c.json
- … and 54 more
Open the folder on GitHubat commit a38f5fc
Compare with similar skills
Mark Zuckerberg 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 |
|---|---|---|---|---|---|---|
| Mark Zuckerberg this skillK-Dense-AI/mimeographs | 129 | — | ~1.4k | Automated safety check: Pass | MIT | |
| Karpathy Guidelinesalirezarezvani/ClaudeForge | 429 | — | ~1.2k | Automated safety check: Pass | MIT | |
| Auto HarnessPacificStudio/openase | 268 | — | ~1.2k | Automated safety check: Pass | Apache-2.0 | |
| Existing Repoalinaqi/maggy | 707 | — | ~4k | Automated safety check: Notes | MIT | |
| Figma Policy Guardrailsjeremylongshore/tons-of-skills-marketplace | 2.8k | — | ~2k | Automated safety check: Notes | MIT | |
| Common Agent GuardrailsHoangNguyen0403/agent-skills-standard | 572 | — | ~1k | Automated safety check: Notes | MIT |
Similar skills
Auto Harness
PacificStudio/openase
Diagnose and strengthen a repository's harness layer: AGENTS.md rules, knowledge layout, architecture boundaries, lint and type gates, API and generated-client contracts, test scaffolding…
Existing Repo
alinaqi/maggy
Analyze existing repositories, maintain structure, setup guardrails and best practices
Figma Policy Guardrails
jeremylongshore/tons-of-skills-marketplace
Enforce security policies and coding standards for Figma API integrations.
Common Agent Guardrails
HoangNguyen0403/agent-skills-standard
Define deterministic guardrails for agent tool calls — protected paths, test-file locks during bug fixes, post-edit formatters, production approval gates, secret deny rules.
Git Guardrails Claude Code
fossasia/eventyay-interpretation
Set up Claude Code hooks to block dangerous git commands (push, reset --hard, clean, branch -D, etc.) before they execute.
More from K-Dense-AI/mimeographs
All 60 skills in this repoAlbert Hofman
K-Dense-AI/mimeographs
Applies the epidemiological reasoning and population-health frameworks of Albert Hofman (Harvard epidemiologist, Rotterdam Study).
Andrew Carnegie
K-Dense-AI/mimeographs
Applies the strategic, philanthropic, and operational frameworks of Andrew Carnegie, founder of Carnegie Steel.
Anne Wojcicki
K-Dense-AI/mimeographs
Applies the strategic frameworks and mental models of Anne Wojcicki, co-founder and CEO of 23andMe.
Aristotle
K-Dense-AI/mimeographs
Applies the frameworks of Aristotle (ancient Greek philosopher, logic, ethics, metaphysics, 384-322 BCE) to decision-making, ethics, and analysis.
Aviv Regev
K-Dense-AI/mimeographs
Applies the computational biology and AI-driven reasoning of Aviv Regev (computational biologist, Genentech, single-cell genomics).
Bill Gates
K-Dense-AI/mimeographs
Apply the mental models of Bill Gates, co-founder of Microsoft and philanthropist.
Related
Categories
Questions about Mark Zuckerberg
What does Mark Zuckerberg do?
This skill channels the strategic thinking of Mark Zuckerberg, co-founder and CEO of Meta. Mark Zuckerberg is an agent skill from K-Dense-AI/mimeographs. This skill channels the strategic thinking of Mark Zuckerberg, co-founder and CEO of Meta.
When should I use Mark Zuckerberg?
Mark Zuckerberg fits situations like: apply his frameworks on decentralized testing; small group context sharing; progressive AI integration; even if the user doesnt explicitly mention his name.
How do I install Mark Zuckerberg in Claude Code?
Run `npx skills add K-Dense-AI/mimeographs --skill mark-zuckerberg -a claude-code`. Or copy the skill folder (mimeographs/mark-zuckerberg in K-Dense-AI/mimeographs) into .claude/skills/mark-zuckerberg in your project. Claude Code loads it when a task matches its description.
How do I install Mark Zuckerberg in Codex?
Run `npx skills add K-Dense-AI/mimeographs --skill mark-zuckerberg -a codex`. Or copy the skill folder (mimeographs/mark-zuckerberg in K-Dense-AI/mimeographs) into .agents/skills/mark-zuckerberg in your project. Codex loads it when a task matches its description.
Can I use Mark Zuckerberg 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 K-Dense-AI/mimeographs --skill mark-zuckerberg -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/mark-zuckerberg, .gemini/skills/mark-zuckerberg, .github/skills/mark-zuckerberg and .opencode/skills/mark-zuckerberg in your project.
What does Mark Zuckerberg need to run?
SKILL.md names no scripts, command-line tools or credentials: Mark Zuckerberg is instructions for the agent only.
Does Mark Zuckerberg access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Mark Zuckerberg 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 Mark Zuckerberg use?
Mark Zuckerberg 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 Mark Zuckerberg use?
About 1.4k tokens (SKILL.md is roughly 5.5k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 4.1k tokens, read only when the agent opens those files.
What are the alternatives to Mark Zuckerberg?
Skills that share tags, products or a category with Mark Zuckerberg: Karpathy Guidelines (alirezarezvani/ClaudeForge, 429 stars), Auto Harness (PacificStudio/openase, 268 stars), Existing Repo (alinaqi/maggy, 707 stars) and Figma Policy Guardrails (jeremylongshore/tons-of-skills-marketplace, 2.8k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Mark Zuckerberg?
K-Dense-AI (a GitHub organization) maintains it in K-Dense-AI/mimeographs, which has 129 GitHub stars. The repository holds 60 skills in this directory. The repository was last updated on August 18, 2026.
Source: K-Dense-AI/mimeographs on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.