skills.rest
andregusman-raiz
andregusman-raizCOMMUNITY
Explore all skills by this author →
GitHub

ag-criar-cenario-benchmark

Create TypeScript benchmark scenario files with L1-L4 criteria for QAT testing.

19|4|Updated Mar 7, 2026
Search Tags:#benchmark#scenarios#dimensions#scenario-design#dual-run#qat#anticontamination
One-click install
npx skills add https://github.com/andregusman-raiz/a-gusman-claude --skill ag-criar-cenario-benchmark
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ag-criar-cenario-benchmark
Source: https://github.com/andregusman-raiz/a-gusman-claude/tree/main/skills/ag-criar-cenario-benchmark
Command: npx skills add https://github.com/andregusman-raiz/a-gusman-claude --skill ag-criar-cenario-benchmark

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The designer creates high-quality benchmark scenarios for QAT benchmarking against a baseline, enabling repeatable evaluation of AI agents on defined criteria.

Core Features & Use Cases

  • Scenario design for eight-dimension benchmarks with dual-run setups, anti-contamination rules, and L1-L4 criteria per dimension.
  • Outputs TypeScript scenario files in scenarios/fixed and scenarios/rotatable with IDs, prompts, target dimensions, and functional checks.
  • Interacts with ag-benchmark-qualidade to execute benchmarks and compare results.

Quick Start

Create a set of rotatable benchmark scenarios for ag-criar-cenario-benchmark focusing on tool use.

Frequently Asked Questions about ag-criar-cenario-benchmark

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

FAQPage Schema
How do I design benchmark scenarios for evaluating AI agents against a baseline?▼

To design benchmark scenarios for evaluating AI agents, you create structured TypeScript files covering eight dimensions with dual-run setups and L1-L4 criteria, enabling repeatable comparative analysis against a defined baseline.

What are dual-run setups and anti-contamination rules in benchmark scenario design?▼

Dual-run setups and anti-contamination rules in benchmark scenario design prevent evaluation interference by establishing controlled execution parameters, ensuring comparative analysis remains isolated and repeatable across the eight defined dimensions.

How do I structure TypeScript benchmark scenario files for rotatable and fixed dimensions?▼

You structure TypeScript benchmark scenario files by placing them under scenarios/fixed and scenarios/rotatable directories, including scenario IDs, prompts, target dimensions, and functional checks for execution.

Can I use L1-L4 criteria to evaluate specific dimensions like tool use in QAT benchmarking?▼

Yes, you can use L1-L4 criteria to evaluate specific dimensions like tool use in QAT benchmarking. The framework applies these criteria per dimension to generate rotatable benchmark scenarios for comparative analysis.

Does this benchmark scenario generator integrate with other tools to execute and compare results?▼

Yes, the benchmark scenario generator integrates with ag-benchmark-qualidade to execute benchmarks and compare results, completing the workflow from scenario creation to final evaluation.

Related Skills

Explore More Education & Research Skills→
ClyptAIClyptAI

benchmark

Measure real-browser performance by collecting navigation and resource metrics for target URLs.

Official
Advanced
matantsachmatantsach
1

run-evals

Run existing skill evaluations via npx snapeval eval and parse benchmark results.

Community
Intermediate
andregusman-raizandregusman-raiz
19

ag-criar-cenario-qat

Translate user stories into four-layer QAT specifications with golden samples and anti-patterns.

Community
Advanced
danielscholldanielscholl
23

test-framework

Validate plugin skills across four layers with dry-run and JSON CI outputs.

Community
Advanced