skill-authoring

Create, edit, and validate SLICC agent skills with frontmatter and companion files.

28|14|Updated Mar 2, 2026
One-click install
npx skills add https://github.com/ai-ecoverse/slicc --skill skill-authoring-ai-ecoverse
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-authoring
Source: https://github.com/ai-ecoverse/slicc/tree/main/packages/vfs-root/workspace/skills/skill-authoring
Command: npx skills add https://github.com/ai-ecoverse/slicc --skill skill-authoring-ai-ecoverse

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill eliminates the guesswork and common pitfalls of writing SLICC agent skills, solving the problem of skills that fail to load, have broken trigger descriptions, or don't follow platform conventions that prevent them from working as intended.

Core Features & Use Cases

  • Frontmatter & Discovery Rules: Covers required SKILL.md fields (name, description, allowed-tools), valid skill root paths (native /workspace/skills/ vs compatibility .agents/ and .claude/ roots), and discovery precedence to ensure skills are found by the agent.
  • Trigger Description Best Practices: Teaches how to write description fields that reliably match user intent, avoiding vague keyword lists that cause skills to load at irrelevant times.
  • Companion File Guidance: Explains when and how to ship .jsh JavaScript shell scripts for deterministic tasks and .bsh browser hooks for site-specific automations, including runtime surface rules and capability bridges.
  • Use Case Example: If you want to build a skill that automates pulling invoice data from a vendor portal, this Skill teaches you how to structure the SKILL.md frontmatter so the agent loads it only when the user asks for invoice-related tasks, and how to add a .jsh helper for safe, deterministic API calls.

Quick Start

Use the skill-authoring skill to create a new SLICC skill that automates sending daily status updates to your team's Slack channel, following all frontmatter and description best practices to ensure it loads reliably when you ask for Slack-related tasks.

Frequently Asked Questions about skill-authoring

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

FAQPage Schema
How do I write SLICC agent skills that load correctly when triggered?

To write SLICC agent skills that load correctly, you must configure the SKILL.md frontmatter with valid fields and write trigger descriptions that precisely match user intent. This ensures the platform runtime discovers and executes the skill reliably.

What is the correct skill root path for native and compatibility discovery in SLICC?

The correct skill root path for SLICC discovery is `/workspace/skills/` for native skills, while compatibility roots use `.agents/` and `.claude/` directories. Using these valid paths ensures the agent runtime finds your skill during intent matching.

How do I create deterministic tasks using .jsh scripts and browser hooks for agent skills?

To create deterministic tasks for agent skills, you ship companion `.jsh` JavaScript shell scripts for safe API calls and `.bsh` browser hooks for site-specific automations. These files integrate with the platform's runtime surface for deterministic execution.

Why does my agent skill fail to load when users ask for relevant tasks?

Your agent skill fails to load when its trigger description uses vague keyword lists instead of specific intent matching. Refining the SKILL.md frontmatter description to accurately reflect the task context prevents it from loading at irrelevant times.

What are the required frontmatter fields for a valid SKILL.md file?

The required frontmatter fields for a valid SKILL.md file are `name`, `description`, and `allowed-tools`. Configuring these fields correctly ensures your skill follows platform conventions and is discovered by the agent runtime surface.