Skill

Design, document, and deploy custom AI skills to spec

Meta-skill that designs, writes, and registers new skills into the Antigravity repository following its Diamond Standard.

Works with github

77
Spark score
out of 100
Updated 29 days ago
Version 14.1.0

Add to Favorites

Why it matters

Automate the end-to-end creation of production-ready AI skills that follow organizational standards, from gathering requirements through code generation to repository integration.

Outcomes

What it gets done

01

Collect skill DNA: technical name, expert role, and key outputs

02

Generate structured README documentation with capabilities and usage rules

03

Create registration snippets for skill orchestrator integration

04

Deploy skill folders and update master registry in the repository

Install

Add it to your toolbox

Run in your project directory:

curl -fsSL https://spark.entire.vc/get/ag-10-andruia-skill-smith | bash

Overview

Andru.ia Skill-Smith (The Forge)

A meta-skill that forges new skills for the Antigravity repository through a three-phase protocol - DNA definition, README/registry generation, and folder deployment - following the repository's Diamond Standard. Use it whenever a new skill needs to be added to the Antigravity repository with consistent structure, Spanish documentation, and a defined expert role.

What it does

Andru.ia Skill-Smith ("The Forge") acts as Andru.ia's Systems Engineer, designing, drafting, and deploying new skills into the repository while ensuring they follow the official Antigravity folder structure and Diamond Standard. Every skill it generates must have its instructions and documentation written in Spanish, follow the formal anatomy (folder -> README.md -> registry entry), and have a defined expert role rather than being generic.

When to use - and when NOT to

Use it when the workflow calls for creating a new skill within the Antigravity repository following its official structure and quality standard.

Inputs and outputs

A three-phase forging protocol: Phase 1, Skill DNA, asks the user for three pillars - a technical name (e.g. @cyber-sec, @data-visualizer), an expert role (who this AI is, e.g. "a security audit expert"), and key outputs (what specific files or actions it performs). Phase 2, Materialization, generates a customized README.md (description, capabilities, golden rules, usage mode) and a ready-to-insert registry snippet for the repository's "Full skill registry" table. Phase 3, Deployment and Integration, creates the physical folder under the skills directory, writes the README.md into it, and updates the repository's master registry so the orchestrator recognizes the new skill.

Integrations

Two golden rules govern every generated skill: numeric prefixes (a correlative number assigned to the folder, e.g. 11, 12, 13, to preserve ordering) and prompt engineering (instructions must include few-shot or chain-of-thought techniques for maximum precision). It writes directly into the Antigravity repository's skill registry and folder structure so newly forged skills are immediately discoverable by the orchestrator.

Who it's for

Maintainers of the Antigravity skills repository who need new skills created to a consistent, senior-quality standard - Spanish documentation, defined expert role, proper folder/registry structure - rather than ad hoc, inconsistently structured additions.

FAQ

Common questions

Discussion

Questions & comments · 0

Sign In Sign in to leave a comment.