Skip to content

Latest commit

 

History

History
82 lines (54 loc) · 3.2 KB

VSCODE.md

File metadata and controls

82 lines (54 loc) · 3.2 KB

VSCODE.md

This file provides guidance for using Skiller with GitHub Copilot in Visual Studio Code.

Project Overview

Skiller is an agent skill for continuous learning — it enables AI coding agents to autonomously extract and preserve learned knowledge into reusable skills. It is not an application codebase but rather a skill definition with documentation and examples.

Key Files

  • SKILL.md — The main skill definition (YAML frontmatter + instructions). This is what agents load.
  • .github/agents/skiller.agent.md — Custom Copilot agent for skill extraction
  • .github/prompts/skiller.prompt.md — Prompt file for /skiller command in Copilot Chat
  • .github/copilot-instructions.md — Repository-level custom instructions
  • resources/skill-template.md — Template for creating new skills

VS Code Integration

Custom Agent (@skiller)

Copilot detects .agent.md files in .github/agents/ automatically. Type @skiller in Copilot Chat to invoke the Skiller agent for skill extraction.

Prompt File (/skiller)

Copilot detects .prompt.md files in .github/prompts/ automatically. Type /skiller in Copilot Chat to trigger a session retrospective.

Agent Skills (Automatic)

Skills in .github/skills/ are discovered automatically by Copilot's progressive loading system. When your current context matches a skill's description, Copilot loads it.

Custom Instructions

The .github/copilot-instructions.md file provides repository-level instructions that Copilot applies to all interactions in this workspace.

How Skills Load in VS Code

Copilot uses three-level progressive loading:

  1. Discovery: Name + description loaded at startup (~100 tokens per skill)
  2. Activation: Full SKILL.md loaded when request matches description
  3. Execution: Supporting files loaded only as needed

Skill File Format

Skills use YAML frontmatter followed by markdown:

---
name: kebab-case-name
description: |
  Must be precise for semantic matching. Include:
  (1) exact use cases, (2) trigger conditions like error messages,
  (3) what problem this solves
author: Skiller
version: 1.0.0
date: YYYY-MM-DD
---

The description field is critical — it determines when the skill surfaces during semantic matching.

Installation Paths

  • Project-level (recommended): .github/skills/[skill-name]/SKILL.md
  • User-level: ~/.copilot/skills/[skill-name]/SKILL.md

The .github/skills/ path is the cross-agent standard and is recommended for maximum compatibility.

Quality Criteria for Skills

When modifying or creating skills, ensure:

  • Reusable: Helps with future tasks, not just one instance
  • Non-trivial: Requires discovery, not just documentation lookup
  • Specific: Clear trigger conditions (exact error messages, symptoms)
  • Verified: Solution has actually been tested and works

References