v05-eval

Automate V05 outfit recommendation evaluation with live scenarios and log mining.

1|Updated May 4, 2026

Search Tags:#qa #evaluation #database analysis #outfit recommendation #v05

Ready to run

npx skills add https://github.com/auxi-wardrobe/auxi-all-in --skill v05-eval

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: v05-eval
Source: https://github.com/auxi-wardrobe/auxi-all-in/tree/main/.agents/skills/v05-eval
Command: npx skills add https://github.com/auxi-wardrobe/auxi-all-in --skill v05-eval

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires requests, pandas, sqlalchemy, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates the evaluation of outfit recommendations for the V05 engine, streamlining the QA process and identifying potential issues in advance.

Core Features & Use Cases

End-to-End QA Harness: Replaces manual evaluation workflows with a single command.
Live Scenario Execution: Runs evaluation scenarios for real-time feedback.
DB Log Mining: Analyzes logs for failure patterns and potential regressions.
Multimodal Scoring: Provides comprehensive scoring based on various criteria.
Use Case: Before shipping a new V05 feature, use this Skill to ensure the outfit recommendations meet quality standards and do not contain regressions.

Quick Start

Run the v05-eval skill with the --fresh flag to execute live evaluation scenarios.

Frequently Asked Questions about v05-eval

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

FAQPage Schema

How do I automate outfit recommendation evaluation for my recommendation engine?▼

Automating outfit recommendation evaluation involves running live scenario execution, mining database logs for failure patterns, and applying multimodal scoring to ensure quality. This Skill orchestrates the entire QA harness using Python to replace manual workflows.

How does database log mining work for finding regressions in outfit recommendations?▼

Database log mining for finding regressions works by analyzing backend logs to extract failure patterns and identify potential issues in outfit recommendations. This Skill uses sqlalchemy and pandas to query and process logs for regression detection.

Do I need access to the wardrobe-backend to run live evaluation scenarios?▼

Yes, you need access to the wardrobe-backend and database to run live evaluation scenarios. The Skill requires this connectivity to execute real-time feedback scenarios and mine logs for the V05 outfit recommendation engine.

What is multimodal scoring in the context of outfit recommendation QA?▼

Multimodal scoring in outfit recommendation QA is the process of providing comprehensive evaluations based on various criteria. This Skill calculates these scores automatically after executing live scenarios to ensure recommendations meet quality standards.

How do I execute a fresh evaluation run for outfit recommendations?▼

To execute a fresh evaluation run for outfit recommendations, use the --fresh flag with the Skill. This triggers live scenario execution to provide real-time feedback on the V05 engine's current outfit recommendation quality.