Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add alirezarezvani/ClaudeForge --skill karpathy-guidelinesgit clone --depth 1 https://github.com/alirezarezvani/ClaudeForgeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/alirezarezvani/claudeforge/karpathy-guidelines)<a href="https://agentmods.dev/skills/alirezarezvani/claudeforge/karpathy-guidelines"><img src="https://agentmods.dev/badge/skills/alirezarezvani/claudeforge/karpathy-guidelines.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00050 | $0.01132 |
| Opus 5 | $0.00025 | $0.00566 |
| Sonnet 5 | $0.00010 | $0.00226 |
| Haiku 4.5 | $0.00005 | $0.00113 |
Grade A, and why
karpathy-guidelines scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 130 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Karpathy Guidelines for LLM Coding
Behavioral guardrails for code generation in Claude Code projects, distilled from observations on common LLM coding failure modes. Apply these to every editing, reviewing, and refactoring task.
Attribution: adapted from the MIT-licensed
karpathy-guidelinesskill by Forrest Chang (https://github.com/forrestchang/andrej-karpathy-skills), inspired by Andrej Karpathy's commentary on where LLM-generated code typically goes wrong. ClaudeForge integrates these principles so every project initialised or enhanced through/enhance-claude-mdships with them in its CLAUDE.md.
When to Apply
Apply on every non-trivial task: writing new code, editing existing code, code review, refactoring, and bug fixing. They are intentionally conservative — bias toward caution over speed.
1. Think Before Coding
Surface what is uncertain. Do not paper over confusion with plausible-sounding code.
- State assumptions before implementing; if any are load-bearing and you are not sure, ask.
- If the request admits more than one reasonable interpretation, list them rather than silently picking one.
- If a simpler approach exists than the one the user proposed, say so and explain the tradeoff.
- When something is genuinely unclear, stop. Identify what is unclear in concrete terms, then ask.
2. Simplicity First
Write the minimum code that solves the stated problem. Nothing speculative.
- Do not add features that were not requested.
- Do not introduce abstractions when there is only one call site.
- Do not add configuration knobs or extension points on speculation.
- Do not add error handling for conditions that cannot occur in this code path.
- If the first draft is 200 lines and a 50-line version would do, rewrite it before shipping.
Self-check: a senior engineer skimming this diff — would they say it is overcomplicated for what was asked? If yes, simplify.
3. Surgical Changes
Touch only what the task requires. Do not opportunistically refactor.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 8d ago First seen · 130 lines · 50 tokens per session scan A acd92adddfb6
karpathy-guidelines is a skill published in the GitHub repository alirezarezvani/ClaudeForge (424 stars, last pushed 3mo ago), licensed MIT. It adds 50 tokens to every session and 1,132 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
mcp-host-styling-integration
Integrates MCP App UI with host theming system. Applies host CSS variables, handles onhostcontextchanged, safe area insets, display mode detection, and fullscreen configuration.
cog-knowledge-consolidation
Build structured knowledge frameworks from scattered vault notes with source attribution.
quality-hooks
Language-specific auto-lint/format/typecheck pipeline. Supports Python (ruff+pyright), TypeScript (prettier+eslint+tsc), Go (gofmt+golangci-lint). Auto-fix and convergence loops.
moai-workflow-thinking
Sequential Thinking MCP for structured step-by-step analysis via --deepthink flag. Separate from UltraThink which is Claude's native extended reasoning mode. Use for multi-step analysis or architecture decisions.
context-engineering
Dynamic context injection, mode switching (dev/review/research), selective loading, and strategic compaction for token optimization.
strict-tdd
Strict RED->GREEN->REFACTOR test-driven development with enforcement. Never write production code before a failing test. Atomic commits per TDD cycle.