Add opt-in Karpathy guidelines skill (#909)

* Add opt-in Karpathy guidelines skill

* Potential fix for pull request finding

Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>

---------

Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
This commit is contained in:
TechBrewBoss
2026-05-04 20:55:05 +08:00
committed by GitHub
co-authored by Copilot Autofix powered by AI
parent de0e395467
commit 6636bce74b
3 changed files with 142 additions and 4 deletions
+5 -4
View File
@@ -10,14 +10,15 @@
* claude-in-chrome), use src/skills/bundled/ instead.
*
* To add a new built-in plugin:
* 1. Import registerBuiltinPlugin from '../builtinPlugins.js'
* 2. Call registerBuiltinPlugin() with the plugin definition here
* 1. Import its registerXPlugin() helper from the plugin module here
* 2. Call that helper from initBuiltinPlugins()
*/
import { registerKarpathyGuidelinesPlugin } from './karpathyGuidelines.js'
/**
* Initialize built-in plugins. Called during CLI startup.
*/
export function initBuiltinPlugins(): void {
// No built-in plugins registered yet — this is the scaffolding for
// migrating bundled skills that should be user-toggleable.
registerKarpathyGuidelinesPlugin()
}
@@ -0,0 +1,41 @@
import { afterEach, expect, test } from 'bun:test'
import {
clearBuiltinPlugins,
getBuiltinPluginDefinition,
} from '../builtinPlugins.js'
import { registerKarpathyGuidelinesPlugin } from './karpathyGuidelines.js'
afterEach(() => {
clearBuiltinPlugins()
})
test('karpathy guidelines registers as an opt-in built-in plugin', () => {
registerKarpathyGuidelinesPlugin()
const plugin = getBuiltinPluginDefinition('karpathy-guidelines')
expect(plugin).toBeDefined()
expect(plugin?.defaultEnabled).toBe(false)
expect(plugin?.skills?.map(skill => skill.name)).toEqual([
'karpathy-guidelines',
])
})
test('karpathy guidelines skill includes optional user focus', async () => {
registerKarpathyGuidelinesPlugin()
const skill = getBuiltinPluginDefinition('karpathy-guidelines')?.skills?.[0]
expect(skill).toBeDefined()
const blocks = await skill!.getPromptForCommand(
'prefer tests over snapshots',
{} as never,
)
const text = (blocks[0] as { text: string }).text
expect(text).toContain('# CLAUDE.md')
expect(text).toContain('These guidelines are working if')
expect(text).toContain('## User Focus')
expect(text).toContain('prefer tests over snapshots')
})
+96
View File
@@ -0,0 +1,96 @@
import { registerBuiltinPlugin } from '../builtinPlugins.js'
const KARPATHY_GUIDELINES_PROMPT = `# CLAUDE.md
Behavioral guidelines to reduce common LLM coding mistakes. Merge with project-specific instructions as needed.
**Tradeoff:** These guidelines bias toward caution over speed. For trivial tasks, 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"
For multi-step tasks, state a brief plan:
\`\`\`
1. [Step] -> verify: [check]
2. [Step] -> verify: [check]
3. [Step] -> verify: [check]
\`\`\`
Strong success criteria let you loop independently. Weak criteria ("make it work") require constant clarification.
---
**These guidelines are working if:** fewer unnecessary changes in diffs, fewer rewrites due to overcomplication, and clarifying questions come before implementation rather than after mistakes.
`
export function registerKarpathyGuidelinesPlugin(): void {
registerBuiltinPlugin({
name: 'karpathy-guidelines',
description:
'Optional coding guidelines that favor simple, surgical, verifiable changes.',
version: '1.0.0',
defaultEnabled: false,
skills: [
{
name: 'karpathy-guidelines',
description:
'Apply coding guidelines that reduce common LLM implementation mistakes.',
whenToUse:
'Use when writing, reviewing, or refactoring code, especially when a task could become overcomplicated or needs careful verification.',
userInvocable: true,
async getPromptForCommand(args) {
const focus = args.trim()
const prompt = focus
? `${KARPATHY_GUIDELINES_PROMPT}\n\n## User Focus\n\n${focus}\n`
: KARPATHY_GUIDELINES_PROMPT
return [{ type: 'text', text: prompt }]
},
},
],
})
}