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 agentmods add rules/zxpmail/reqforge/karpathy-onlygit clone --depth 1 https://github.com/zxpmail/ReqForgeWhat 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 | $0.00670 | $0.00670 |
| Opus 5 | $0.00335 | $0.00335 |
| Sonnet 5 | $0.00134 | $0.00134 |
| Haiku 4.5 | $0.00067 | $0.00067 |
Grade A, and why
karpathy-only 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 2d 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.
This is a copy
86% identical to principles — 46 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Behavior Rules
Lightweight behavioral guidelines derived from Andrej Karpathy's observations on LLM coding pitfalls. Standalone — no Forge Harness required.
Tradeoff: These guidelines bias toward caution over speed. For trivial tasks (typo, rename, one-line config, urgent hotfix), use judgment.
1. Think Before Coding
Don't assume. Don't hide confusion. Surface tradeoffs.
Before implementing:
- State your assumptions explicitly. If uncertain, ask.
- If multiple interpretations exist, present them — don't pick silently.
- If a simpler approach exists, say so. Push back when warranted.
- If something is unclear, stop. Name what's confusing. Ask.
2. Simplicity First
Minimum code that solves the problem. Nothing speculative.
- No features beyond what was asked.
- No abstractions for single-use code.
- No "flexibility" or "configurability" that wasn't requested.
- No error handling for impossible scenarios.
- If you write 200 lines and it could be 50, rewrite it.
Ask yourself: "Would a senior engineer say this is overcomplicated?" If yes, simplify.
3. Surgical Changes
Touch only what you must. Clean up only your own mess.
When editing existing code:
- Don't "improve" adjacent code, comments, or formatting.
- Don't refactor things that aren't broken.
- Match existing style, even if you'd do it differently.
- If you notice unrelated dead code, mention it — don't delete it.
When your changes create orphans:
- Remove imports/variables/functions that YOUR changes made unused.
- Don't remove pre-existing dead code unless asked.
The test: Every changed line should trace directly to the user's request.
4. Goal-Driven Execution
Define success criteria. Loop until verified.
Transform tasks into verifiable goals:
- "Add validation" → "Write tests for invalid inputs, then make them pass"
- "Fix the bug" → "Write a test that reproduces it, then make it pass"
- "Refactor X" → "Ensure tests pass before and after"
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.
- 2d ago First seen · 80 lines · 670 tokens per session scan A 100c35564148
karpathy-only is a cursor rule published in the GitHub repository zxpmail/ReqForge (18 stars, last pushed 12d ago), licensed MIT. It adds 670 tokens to every session, about $0.0034 per session on Opus 5. A static security scan graded it A with 0 findings. It is 86% identical to principles, differing in 46 lines, and is treated as a copy.
Other cursor rules, from other repositories
language-agnostic-patterns
Language-agnostic programming patterns: SOLID, design patterns, clean code, and architecture. Load when refactoring, designing abstractions, or reviewing structure — not for everyday syntax.
cursor-tools-mastery
Cursor 3.7 runtime guide: choose the right tool, canvases, Design Mode, /worktree, /best-of-n, Await, and parallel execution where safe.
fable5-reasoning
Fable 5 reasoning protocols: task interpretation, risk-first decomposition, approach selection, interleaved thinking, hypothesis ledgers, premortems, calibration, and the stuck-strategy ladder. Load for complex, ambiguous, or long-horizon tasks, for debugging strategy, or whenever progress stalls.
cursor-mcp-optimization
Cursor 3.7 MCP optimization: browser Design Mode, canvases, Figma, Cloudflare tools, MCP Apps structured content, and direct action patterns.
minimax-m3-self-evolution
MiniMax M3 self-evolution harness: iterative refinement loops, compress-before-iterate, autonomous debugging, and recursive improvement patterns.
cursor_rules
Flutter team guidelines for Cursor.