skills.rest
LiorCohen
LiorCohenCOMMUNITY
Explore all skills by this author →
GitHub

Plugin Testing Standards

Establish tiered Unit, Workflow, and Integration tests for AI code plugins.

39|3|Updated Nov 21, 2025
Search Tags:#code quality#testing#llm#ai plugins#deterministic testing#workflow verification
One-click install
npx skills add https://github.com/LiorCohen/sdd --skill plugin-testing-standards
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: Plugin Testing Standards
Source: https://github.com/LiorCohen/sdd/tree/main/.claude/skills/plugin-testing-standards
Command: npx skills add https://github.com/LiorCohen/sdd --skill plugin-testing-standards

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill provides a comprehensive methodology and set of best practices for testing AI-driven code plugins, ensuring reliability, determinism, and quality in LLM-generated workflows.

Core Features & Use Cases

  • Structured Testing Tiers: Defines Unit, Workflow, and Integration tests for thorough verification.
  • Deterministic LLM Testing: Outlines methods for capturing and parsing LLM output for consistent results.
  • Prompt Engineering Guidelines: Provides essential instructions for automated test prompts to ensure predictable AI behavior.
  • Use Case: A plugin developer needs to ensure their new code generation plugin consistently produces valid, compilable code and correctly invokes specified tools or skills. This Skill's framework guides them through creating tests that verify these aspects rigorously.

Quick Start

Follow the plugin testing standards to create deterministic tests for your AI plugin.

Frequently Asked Questions about Plugin Testing Standards

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

FAQPage Schema
How do I test LLM workflows for deterministic output?▼

Testing AI plugin code requires a tiered approach covering Unit, Workflow, and Integration tests to verify code generation consistency. This framework ensures plugins produce valid, compilable code and correctly invoke specified tools.

What is the best way to structure automated tests for AI plugins?▼

When AI plugin tests fail, you should fix the underlying plugin code rather than modifying the test assertions. This principle ensures the testing framework maintains its integrity and accurately verifies LLM-driven workflows.

How do I capture and parse LLM output during automated testing?▼

Prompt engineering for automated testing involves providing specific instructions to guide the LLM into producing predictable, testable outputs. This ensures the AI plugin consistently generates valid code and correctly invokes specified tools or skills.

Why do my AI plugin tests fail intermittently and how should I fix them?▼

To ensure AI plugins correctly invoke specified tools, implement Integration and Workflow tests within a tiered testing framework. This verifies that LLM-driven workflows execute tool calls accurately and produce valid, compilable code.

Related Skills

Explore More Software Engineering Skills→
1molchuan1molchuan
1

oh-my-ccg-init

Verify MCP, Codex, and Gemini endpoints and initialize RPI state.

Community
Intermediate
HomericIntelligenceHomericIntelligence
18

run-tests

Execute Mojo or Python test suites and generate validation reports.

Official
Basic
sambersamber
2.9k

golang-lint

Configure and run golangci-lint to analyze Go projects and interpret results.

Community
Advanced
qazz92qazz92

test-automation

Automate test generation, coverage analysis, and maintenance for software projects.

Community
Intermediate