skills.rest
udecode
udecodeOFFICIAL
Explore all skills by this author →
WebsiteGitHub

slate-ar-recipe

Recommends Codex Autoresearch recipes and generates read-only setup plans for Slate v2 measurement loops.

16.5k|997|Updated Dec 2, 2019
Search Tags:#CLI#benchmark#Performance#Recipe#AutoResearch#Slate#setup-plan
One-click install
npx skills add https://github.com/udecode/plate --skill slate-ar-recipe
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: slate-ar-recipe
Source: https://github.com/udecode/plate/tree/main/.agents/skills/slate-ar-recipe
Command: npx skills add https://github.com/udecode/plate --skill slate-ar-recipe

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Choosing the right Autoresearch loop for a measured task in the Slate v2 codebase is ambiguous when the work is not obviously performance-related, such as test runtime, typecheck time, bundle size, memory, or command latency.

Core Features & Use Cases

  • Recipe Discovery: Lists, recommends, and shows Codex Autoresearch recipes against the .tmp/slate-v2 working directory.
  • Read-Only Setup Plans: Produces setup-plan output without starting any measurement packets unless explicitly requested.
  • Loop Routing: Maps non-performance loops like node-test-runtime, typescript-compile-time, bundle-size, memory-usage, command-latency, and quality-gap to the right owner skill.
  • Use Case: A developer notices Slate v2 typecheck is slow and asks which Autoresearch loop to run; the skill recommends typescript-compile-time, reports the metric, benchmark command, and hands off to the appropriate owner.

Quick Start

Ask which Autoresearch recipe fits a slow Slate v2 typecheck and request a read-only setup plan for it.

Frequently Asked Questions about slate-ar-recipe

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I choose an Autoresearch recipe for a measured task?▼

Run the recipes recommend command against the target working directory to get a suggested recipe, then use recipes show to inspect its metric, direction, and benchmark command before generating a setup plan.

What Autoresearch recipes exist for non-performance tasks?▼

Built-in recipes cover node-test-runtime, vitest-runtime, typescript-compile-time, bundle-size, memory-usage, command-latency, and quality-gap. A custom recipe is the last resort when no built-in option fits the measured task.

Does the recipe picker start benchmark loops automatically?▼

No, it defaults to read-only operations: recipes list, recommend, show, and setup-plan. Measurement packets only start when the user explicitly asks to run the loop.

When should I use a custom recipe instead of a built-in one?▼

Use custom only when no built-in recipe matches the measured task. For memory-usage and command-latency, you must first replace the placeholder workload or command with the real one before running.

Why prefer explicit Bun commands over the generic npm recipe in Slate v2?▼

In Slate v2, explicit Bun or package commands give more accurate and repeatable measurements than a generic npm recipe when the recommender selects node-test-runtime, so the skill routes those cases accordingly.

Related Skills

Explore More Software Engineering Skills→
kevindreekevindree

performance-optimization

Measure application performance bottlenecks and verify fixes with metrics and budgets.

Community
Advanced
AradotsoAradotso
70

pi-autoresearch-loop

Automates an iterative experiment loop to optimize code and benchmarks.

Official
Advanced
Just-itJust-it
33

ascendc-evaluation

Evaluate AscendC custom operators against reference implementations for correctness and performance.

Community
Advanced
TheRealSeberTheRealSeber
1

jade-recipe-noop

Mark JADE migration findings as SKIPPED or DEFERRED without source transforms.

Community
Intermediate