skill-factory

Automate Claude Code skill creation from research to validated SKILL.md artifacts.

5|Updated Nov 21, 2025
One-click install
npx skills add https://github.com/basher83/agent-auditor --skill skill-factory-basher83
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-factory
Source: https://github.com/basher83/agent-auditor/tree/main/skills/skill-factory
Command: npx skills add https://github.com/basher83/agent-auditor --skill skill-factory-basher83

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill Factory provides a guided, research-backed workflow to create Claude Code skills from idea to validated production-ready artifacts, including automated research gathering, multi-tier validation, and auditing.

Core Features & Use Cases

  • Automated research: Gather domain knowledge using firecrawl API.
  • Structured frontmatter generation: Generate SKILL.md with YAML frontmatter and body.
  • End-to-end validation: Run quick_validate.py, perform runtime loading checks, integration checks, and non-blocking audit.

Quick Start

Run the full workflow with a new skill name. If you have existing research at docs/research/skills/<skill-name>, skip the research phase by passing the path. For targeted fixes, use individual commands such as /meta-claude:skill:research, /meta-claude:skill:create, /meta-claude:skill:review-content, and /meta-claude:skill:validate-audit.

Frequently Asked Questions about skill-factory

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

FAQPage Schema
How do I automate skill creation from research to production-ready artifacts?

Skill Factory automates the entire workflow from initial idea through validated production deployment. It gathers domain research using firecrawl, generates SKILL.md with YAML frontmatter, runs multi-tier validation including runtime checks, and produces audit-ready artifacts meeting Claude Code specifications.

Can I skip the research phase if I already have domain knowledge gathered?

Yes. Pass the path to existing research at docs/research/skills/<skill-name> to skip automated research gathering and proceed directly to structured frontmatter generation, content review, and validation phases.

What validation does the Skill Factory perform before production deployment?

The Skill Factory runs quick_validate.py, performs runtime loading checks to ensure SKILL.md loads correctly, executes integration checks, and conducts a non-blocking audit to confirm compliance with Claude Code specifications and audit-ready standards.

Can I run individual validation steps instead of the full workflow?

Yes. Use targeted commands like `/meta-claude:skill:research`, `/meta-claude:skill:create`, `/meta-claude:skill:review-content`, and `/meta-claude:skill:validate-audit` to fix specific phases without re-running the entire pipeline.

Does Skill Factory handle tiered error handling and compliance checks automatically?

Yes. The workflow includes multi-phase validation, tiered error handling, and compliance checks built into each stage to ensure artifacts meet Claude Code specifications before final audit.