cyberneticlibrary

Run daily multi-tier model performance arena

model-arena-dailyworkflowsetup L30
frankxai/workflow-tier-plugin
What it does

Daily multi-tier model arena. Runs the same

Best for

Benchmarking multi-tier LLM responses against a canonical prompt with regression detection and cost-efficiency scoring.

Inputs
  • · args
  • · args.prompt
  • · args.topic
Outputs
  • · prompt
  • · winner
  • · date
  • · ranking
  • · topicIndex
Requires
  • · agent()
  • · parallel()
Preconditions
  • · Daily intelligence on which Claude tier is best for each task type. Cost-disciplined by design: 3 ge
Failure modes
  • · Agent execution may fail; handled with .catch()
Trust signals
  • · 1 structured output schema(s) defined with required fields
  • · Public GitHub repo: frankxai/workflow-tier-plugin
  • · Canonical meta export with phases and description
  • · Explicit phase() calls (4) for workflow transparency