skill-forge

Validate SKILL.md frontmatter and generate a metadata block with schema.

Updated Feb 12, 2026
One-click install
npx skills add https://github.com/arendon1/agent-skills --skill skill-forge-arendon1
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-forge
Source: https://github.com/arendon1/agent-skills/tree/main/skill-forge
Command: npx skills add https://github.com/arendon1/agent-skills --skill skill-forge-arendon1

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

The Skill Forge provides a structured, auditable approach to creating, testing, and deploying AI agent skills, reducing context rot and increasing reliability.

Core Features & Use Cases

  • Token Loading Hierarchy: store essential metadata globally, load full SKILL.md only when needed, and fetch references on demand.
  • Deterministic tooling: scripts handle heavy logic to ensure reproducibility and reduce model drift.
  • Audit-ready workflows: built-in templates for skill creation, testing, and security audits.
  • Use Case: a team wants to rapidly create, test, and deploy reliable skills with traceable governance.

Quick Start

Start by running the scaffolding tool to bootstrap a new skill and then polish its SKILL.md description for trigger clarity.

Frequently Asked Questions about skill-forge

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

FAQPage Schema
How do I build and audit AI agent skills to reduce model drift?

You can build and audit AI agent skills by using a deterministic framework that validates SKILL.md frontmatter, ensures required resources are present, and relies on scripts for heavy logic to reduce model drift and context rot.

What is a token loading hierarchy for AI skill frameworks?

A token loading hierarchy stores essential metadata globally, loads the full SKILL.md file only when needed, and fetches references on demand to minimize context usage and increase reliability during skill execution.

How do I create a new AI skill with a valid SKILL.md frontmatter schema?

You create a new AI skill by running a scaffolding tool to bootstrap the structure, then defining a SKILL.md file with validated frontmatter including a name, description, and clearly described trigger conditions.

Does this skill framework require dependencies to validate skill metadata?

No, the framework operates with zero external dependencies, using built-in scripts and templates to validate self-contained skill units and produce YAML-ready metadata blocks for audit-ready workflows.

What's the best way to document trigger conditions for LLM automation skills?

The best way to document trigger conditions is to polish the SKILL.md description during creation, ensuring the self-contained skill unit clearly describes when the LLM should activate the automation framework.

Why does my AI skill audit fail when resources are missing?

Your skill audit fails because the validation process checks that required resources like scripts, references, and assets directories are present alongside a valid SKILL.md file to ensure the skill is self-contained.