sdd-forge-skill

Orchestrate the full lifecycle of AI agent skills with creation, evaluation, and optimization modes.

Updated Jan 19, 2026
One-click install
npx skills add https://github.com/sync-dev-org/sync-sdd --skill sdd-forge-skill
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sdd-forge-skill
Source: https://github.com/sync-dev-org/sync-sdd/tree/main/framework/claude/skills/sdd-forge-skill
Command: npx skills add https://github.com/sync-dev-org/sync-sdd --skill sdd-forge-skill

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill streamlines the entire lifecycle of AI agent skills, from initial creation and iteration to rigorous evaluation and optimization, ensuring high-quality, reliable, and well-triggered skills.

Core Features & Use Cases

  • Skill Creation: Guides users through defining and building new skills from scratch.
  • Skill Refinement: Iteratively improves existing skills based on feedback and evaluation results.
  • Evaluation & Benchmarking: Runs automated tests, grades outputs, and provides detailed performance metrics.
  • Comparison: Facilitates blind A/B testing of different skill versions.
  • Description Optimization: Automates the process of improving skill descriptions for better triggering.
  • Use Case: You have an idea for a new skill to help Claude summarize meeting transcripts. Use this Skill to draft the initial SKILL.md, write test cases, evaluate its performance, and refine its description until it reliably triggers when needed.

Quick Start

Use the sdd-forge-skill to create a new skill for summarizing meeting notes.

Frequently Asked Questions about sdd-forge-skill

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

FAQPage Schema
How do I create and iterate on AI agent skills from scratch?

To create AI agent skills, you define the initial concept and draft the SKILL.md file, then use automated testing and iterative feedback loops to refine and improve the skill's performance.

What is the best way to evaluate and benchmark AI prompt performance?

Evaluating AI prompt performance involves running automated tests, grading outputs against defined criteria, and generating detailed performance metrics to ensure high-quality and reliable skill triggering.

How does blind A/B testing work for comparing different skill versions?

Blind A/B testing for skill versions works by executing different implementations of the same prompt concurrently, grading their outputs automatically, and comparing the results to determine which version performs better.

Can I automatically optimize skill descriptions for better AI triggering?

Yes, you can automatically optimize skill descriptions for better AI triggering by using the 'optimize-description' mode, which refines the text to ensure the skill reliably activates when needed.

Does this skill development lifecycle tool require Python for deterministic tasks?

Yes, this skill development lifecycle tool integrates with Python scripts to handle deterministic tasks, ensuring precise execution of automated testing, evaluation routines, and grading logic.

Why should I use iterative development for prompt engineering?

Iterative development for prompt engineering is necessary to systematically refine skills through automated feedback loops, ensuring the final AI agent output is high-quality, reliable, and well-triggered.