Agent-Skills.md

Agent Skills: Benchmark Suite Manager

Manage benchmarks for algorithm engineering experiments and evaluations

research-documentationID: a5c-ai/babysitter/benchmark-suite-manager

Repository

a5c-ai/babysitter
a5c-aiLicense: MIT
244

Install this agent skill to your local

pnpm dlx add-skill https://github.com/a5c-ai/babysitter/tree/HEAD/plugins/babysitter/skills/babysit/process/specializations/domains/science/mathematics/skills/benchmark-suite-manager

Skill Files

Browse the full folder contents for benchmark-suite-manager.

Download Skill

Loading file tree…

plugins/babysitter/skills/babysit/process/specializations/domains/science/mathematics/skills/benchmark-suite-manager/SKILL.md

Skill Metadata

Name
benchmark-suite-manager
Description
Manage and execute mathematical benchmark suites

Benchmark Suite Manager

Purpose

Provides management and execution capabilities for mathematical benchmark suites for algorithm validation.

Capabilities

  • Standard benchmark access (Matrix Market, NIST, etc.)
  • Custom benchmark generation
  • Performance profiling
  • Accuracy validation
  • Comparison against reference solutions
  • Statistical analysis of results

Usage Guidelines

  1. Benchmark Selection: Choose appropriate standard benchmarks
  2. Custom Generation: Create problem-specific benchmarks
  3. Validation: Compare against known solutions
  4. Statistical Analysis: Properly analyze performance data

Tools/Libraries

  • Matrix Market
  • NIST Digital Library
  • SuiteSparse Matrix Collection