skill-test

Official

Streamline Databricks skill testing with YAML GRP

Authordatabricks-solutions
Version1.0.0
Installs0

System Documentation

What problem does it solve?

The Skill Testing Framework centralizes Databricks skill evaluation by providing a reproducible, YAML-first workflow for generating test scaffolding, executing tests (including optional Databricks MCP), and promoting results to ground truth for reliable regression testing.

Core Features & Use Cases

  • YAML-driven test definitions and a Generate-Review-Promote (GRP) pipeline to streamline skill validation.
  • Interactive CLI for creating test prompts, running skills, and reviewing candidates with MLflow-backed metrics.
  • Ground-truth baselines and MLflow integration to track regressions across skill versions and environments.
  • Routing and trace evaluation support to validate skill activation and Claude Code behavior.

Quick Start

  1. Install and configure the skill-test CLI in your project (pip install -e ".test/").
  2. Initialize testing scaffolding for a skill with /skill-test <skill-name> init, then run evaluations with /skill-test <skill-name> run.
  3. Use ground_truth.yaml and baseline.yaml to manage regression tests and track improvements over time.

Dependency Matrix

Required Modules

python-dotenvPyYAMLmlflowdatabricks-sdk

Components

scriptsreferences

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: skill-test
Download link: https://github.com/databricks-solutions/ai-dev-kit/archive/main.zip#skill-test

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 223,000+ vetted skills library on demand.