benchmark-adder

Automate Claude Code plugin creation for benchmarking setups from a repository URL.

3|1|Updated Apr 11, 2026

Search Tags:#benchmarking #Claude Code #plugin creation #benchmark integration #agentic environments

Ready to run

npx skills add https://github.com/tmuskal/arc-agi-benchmarker --skill benchmark-adder

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: benchmark-adder
Source: https://github.com/tmuskal/arc-agi-benchmarker/tree/main/.claude/skills/benchmark-adder
Command: npx skills add https://github.com/tmuskal/arc-agi-benchmarker --skill benchmark-adder

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires babysitter, benchmark-plugin-creator, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates the process of creating Claude Code plugins for benchmarking various setups, allowing users to easily compare their Claude Code configurations against different benchmarks.

Core Features & Use Cases

Plugin Creation: Automatically generates Claude Code plugins for benchmarking different agentic environments, memory/eval benchmarks, QA/code/tool-use benchmarks, etc.
Benchmarking: Orchestrate the creation of a full Claude Code plugin that benchmarks the current harness setup against the specified benchmark.
Integration: Wraps the babysitter:babysit skill with the benchmark-plugin-creator process to integrate the benchmarking functionality into Claude Code.

Quick Start

Run the benchmark-adder skill with the benchmark repository URL you want to create a plugin for.

Frequently Asked Questions about benchmark-adder

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

FAQPage Schema

How do I automate Claude Code plugin creation for benchmarking setups?▼

To automate Claude Code plugin creation for benchmarking, you provide a benchmark repository URL to orchestrate the integration of benchmark plugin creators and babysitter skills, generating a fully configured benchmarking plugin.

What is needed to create a Claude Code plugin for agentic environment benchmarks?▼

Creating a Claude Code plugin for agentic environment benchmarks requires a benchmark repository URL, which the system uses to orchestrate the babysitter and benchmark-plugin-creator processes for automated plugin generation.

Can I benchmark my Claude Code configuration against custom memory and QA benchmarks?▼

Yes, you can benchmark Claude Code configurations against custom memory, QA, code, and tool-use benchmarks by supplying the specific benchmark repository URL to generate the necessary testing plugin.

How does the benchmark integration process work with the babysitter skill?▼

Benchmark integration works by wrapping the babysitter:babysit skill with the benchmark-plugin-creator process, orchestrating both to automatically generate a Claude Code plugin that tests your setup against a specified benchmark.

Are there limitations when automating plugin creation for benchmarking Claude Code?▼

The primary limitation for automating benchmarking plugin creation is the strict requirement for a benchmark repository URL, as the orchestration process cannot generate plugins or integrate benchmarks without this specific input.