karpathy-triplet-diag
CommunityDiagnose loop-readiness by surface, metric, time
Product & Management#metric definition#system diagnostics#autonomous optimization#experimentation design#agent loop readiness#editable surface#time budget
Authordrewid74
Version1.0.0
Installs0
System Documentation
What problem does it solve?
This Skill helps you determine whether a specific system is truly ready for automated optimization by forcing clarity on the editable surface, the optimization metric, and the experiment time budget.
Core Features & Use Cases
- Editable-surface gatekeeping: Identifies what files/configs/prompts/parameters an agent can safely and reversibly change without unpredictable dependency cascades.
- Metric readiness and suitability: Converts “performing well” into an automated, bounded-time, scalar (or weighted scalar) metric with a credible business linkage and sandbox computability.
- Time-budget feasibility: Ensures the system can support hundreds of experiments in an isolated environment within a practical overnight cycle.
Quick Start
Use karpathy-triplet-diag to evaluate a concrete pipeline or codebase by answering the editable-surface, metric, and time-budget questions until the Skill produces either a program.md spec or a blocker report.
Dependency Matrix
Required Modules
None requiredComponents
Standard package💻 Claude Code Installation
Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.
Please help me install this Skill: Name: karpathy-triplet-diag Download link: https://github.com/drewid74/ai_skills/archive/main.zip#karpathy-triplet-diag Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
Agent Skills Search Helper
Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.