skills.rest
Balthael
BalthaelCOMMUNITY
Explore all skills by this author →
GitHub

skill-registry

Generates a skill registry cataloging available skills with compact rules for sub-agent delegation.

Updated Apr 10, 2026
Search Tags:#workflow automation#ai agents#project conventions#agent configuration#skill registry#sub-agent delegation#compact rules
One-click install
npx skills add https://github.com/Balthael/ciberbal-ai --skill skill-registry-balthael
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-registry
Source: https://github.com/Balthael/ciberbal-ai/tree/main/internal/assets/skills/skill-registry
Command: npx skills add https://github.com/Balthael/ciberbal-ai --skill skill-registry-balthael

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? AI agents that delegate work to sub-agents lack a centralized catalog of installed skills and their rules, forcing sub-agents to rediscover conventions on every run. This Skill builds a single registry file with pre-digested compact rules that delegators inject directly into sub-agent prompts. ## Core Features & Use Cases - Skill Scanning: Discovers SKILL.md files across user-level and project-level directories for Claude Code, OpenCode, Gemini CLI, Cursor, VS Code Copilot, and Antigravity, with deduplication and exclusion of SDD workflow skills. - Compact Rules Generation: Distills each skill into 5-15 actionable rules that sub-agents can apply without reading full SKILL.md files. - Convention Indexing: Scans project convention files (AGENTS.md, CLAUDE.md, .cursorrules, GEMINI.md) and extracts all referenced paths so sub-agents need zero extra hops. - Use Case: After installing new skills in a project, run this Skill to regenerate .atl/skill-registry.md so the orchestrator passes accurate, pre-resolved rules to every sub-agent it launches. ## Quick Start Ask the agent to update the skill registry for the current project so it scans installed skills and writes the registry to .atl/skill-registry.md.

Frequently Asked Questions about skill-registry

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

FAQPage Schema
How do I create a skill registry for AI coding agents?▼

Run the skill-registry Skill, which scans all known skill directories for SKILL.md files, extracts names and triggers, generates compact rules per skill, and writes the result to .atl/skill-registry.md in the project root.

How do sub-agents receive skill rules without reading SKILL.md files?▼

The registry contains pre-digested compact rules of 5-15 lines per skill. The delegating agent reads the registry once, resolves matching rules, and injects them directly into each sub-agent's launch prompt.

Which agent skill directories does the registry scan?▼

It scans user-level paths for Claude Code, OpenCode, Gemini CLI, Cursor, and VS Code Copilot, plus project-level paths like .claude/skills, .gemini/skills, .agent/skills, and a generic skills folder, deduplicating in favor of project-level versions.

When should I regenerate the skill registry?▼

Regenerate after installing or removing skills, after setting up a new project, or when explicitly requested. The registry is built once as an expensive operation, then read cheaply at every delegation.

Does the skill registry persist across sessions?▼

Yes. It always writes .atl/skill-registry.md to disk, and if the engram mem_save tool is available, it also saves the registry with a topic_key so repeated runs update the same stored observation.

Related Skills

Explore More Software Engineering Skills→
flekschasflekschas
12

mcp-web-agent-docs

Maintain and update agent documentation and skills in the `./agents/` directory.

Community
Basic
Gentleman-ProgrammingGentleman-Programming
1.2k

skill-registry

Generates a skill registry catalog with compact rules for sub-agent prompt injection.

Official
Intermediate
thiagoooo34252thiagoooo34252
1

sdd-init

Initialize Spec-Driven Development context with tech stack detection and config generation.

Community
Advanced
NUMU-IONUMU-IO
2

speckit-conduct-run

Execute spec-kit phases by parsing instructions and delegating steps to sub-agents.

Official
Advanced