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 jnPiyush/AgentX --skill karpathy-guidelinesgit clone --depth 1 https://github.com/jnPiyush/AgentXWrote 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/jnpiyush/agentx/karpathy-guidelines)<a href="https://agentmods.dev/skills/jnpiyush/agentx/karpathy-guidelines"><img src="https://agentmods.dev/badge/skills/jnpiyush/agentx/karpathy-guidelines/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/jnpiyush/agentx/karpathy-guidelines"><img src="https://agentmods.dev/badge/skills/jnpiyush/agentx/karpathy-guidelines.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Agent Snooping · line 12 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
- medium Excessive Agency · line 166 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00047 | $0.01935 |
| Opus 5.5 | $0.00019 | $0.00774 |
| Sonnet 5.5 | $0.00009 | $0.00387 |
| Haiku 4.5 | $0.00005 | $0.00194 |
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 20d 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 — 178 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Karpathy Guidelines
Behavioral guidelines to reduce common LLM coding mistakes. Adapted from Andrej Karpathy's observations on LLM coding pitfalls and the upstream MIT-licensed karpathy-guidelines skill.
MANDATORY in Frontier: These four guidelines are NON-optional for every Frontier execution (coding, refactor, review, pipeline). Complete the "Self-Check Before Handoff" checklist before any handoff. See the always-on rule in
.github/instructions/project-conventions.instructions.md.
Tradeoff: These guidelines bias toward caution over speed. The only exemption is a genuinely trivial change (single-line typo, obvious config rename) whose success criteria are self-evident.
When to Use
Load this skill when:
- Writing new code or features
- Refactoring or modifying existing code
- Reviewing code (yours or another agent''s)
- Debugging when the cause is not obvious
- Producing pipelines, IaC, or any artifact that ships to production
Rationalization Table
Reject these shortcuts.
| Rationalization | Reality |
|---|---|
| "I understand the intent, I can skip stating assumptions." | You only think you understand. State assumptions explicitly so the user can correct the wrong ones cheaply. |
| "I'll add a small abstraction now, it might be useful later." | Speculative abstractions are the leading source of code that is hard to delete. Inline first, abstract on the second real use. |
| "I'll rewrite this whole function, it will be cleaner." | Rewrites import new bugs and break call sites you did not read. Make the smallest surgical change that satisfies the requirement. |
| "The tests pass, so the change is correct." | Passing tests prove the tests pass. They do not prove the change matches the user's intent or the spec. Re-read the request after the code is written. |
| "I'll add error handling for every imaginable failure." | Defensive code for impossible failures hides the real failures and inflates the diff. Validate at boundaries; trust internal invariants. |
| "I'll polish the comments and structure while I'm here." | Drive-by formatting and comment edits hide the real change from the reviewer. Keep the diff focused; open a separate hygiene PR if needed. |
| "The user did not specify, so I'll pick the safer-sounding option." | Silent picks are silent decisions. Either ask, or pick and surface the decision explicitly so it can be reverted. |
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.
- 20d ago Changed · -4 lines · -14 tokens per session e5844d2071e2
- 1mo ago Changed 6088806bd428
- 1mo ago First seen · 182 lines · 61 tokens per session scan A e938bf4153c4
karpathy-guidelines is a skill published in the GitHub repository jnPiyush/AgentX (16 stars, last pushed 3d ago), licensed Apache-2.0. It adds 47 tokens to every session and 1,935 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-09-03.
Other skills, from other repositories
autoreview
Structured code review through the shared OpenClaw agent-skills installation.
omh-code-review
This is an OMH code-review workflow skill, projected for Agent Skills hosts (Claude Code, Codex, Cursor, opencode, OpenClaw, pi).
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
asplos-writing-style
Use when drafting or revising ASPLOS prose — engineering the first two pages that rapid review actually reads, stating the cross-layer insight as one quotable sentence, structuring per-layer mechanism sections, writing bounded claims, and fitting everything into 11 self-contained pages that include figures and…
explore-unknowns
Guide the user through a quadrant walk that maps the unknowns of a task — open by listing the known knowns, then work through known unknowns, unknown knowns, and unknown unknowns one stage at a time, ending with a complete four-quadrant map in the user's hands. Use when a request is ambiguous or underspecified, the…
re0-merge
Review and land an external contribution the way this suite does: gate it against the thesis, land it with the author's credit intact, complete a new skill rather than merging it raw, then approve, credit, and explain before closing. Use when reviewing a pull request, as any collaborator or maintainer, not only the…