karpathy-planning
Skill2 starsChanged 2 months agoMITView on GitHub
Develop a scoped implementation plan when explicitly asked to plan a coding task. Use grill-me for a deliberate design interview.
In this repository27 other files it gives its agents.
- agent-guardrailsSkill
- benchmark-checklistSkill
- blast-radiusSkill
- blog-reviewSkill
- code-reviewSkill
- create-verification-skillSkill
- datavizSkill
- grill-meSkill
- grok-repoSkill
- hammerspoonSkill
- html-lessonSkill
- mechanism-vizSkill
- obsidian-captureSkill
- obsidian-searchSkill
- paper-searchSkill
- readbackSkill
- repo-evalSkill
- route-skillSkill
- sherlockSkill
- show-meSkill
- sketch-conceptSkill
- skill-curatorSkill
- test-writingSkill
- tla-modelSkill
- web-recapSkill
- write-docsSkill
- write-skillSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 13
Prefer the simplest design that meets the request.
- Distinctive: line 17
If implementation is already authorized, proceed once blocking decisions are resolved.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
karpathy-planning
Produce a plan that establishes scope, important assumptions, and verifiable completion. Include only the detail needed to choose an approach and carry out the task.
- Resolve material ambiguities from available context. Ask when an answer would change the outcome; otherwise state a reasonable assumption.
- Prefer the simplest design that meets the request. Identify existing components to reuse and preserve unrelated work.
- Pair success criteria with relevant verification. Complete required checks; broaden testing only for changes, failures, or unresolved concerns.
- Describe substantial tradeoffs, dependencies, or irreversible steps when they matter.
- Match the output to the task. A four-part template and a repository-wide survey are not prerequisites to coding.
- If implementation is already authorized, proceed once blocking decisions are resolved; planning alone is not completion.
Adapted from multica-ai/andrej-karpathy-skills.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.