copilot-customization
Create, scaffold, and configure GitHub Copilot customizations for VS Code projects. Covers custom instructions (.instructions.md), prompt files (.prompt.md), custom agents (.agent.md), agent skills (SKILL.md folders), hooks (.github/hooks/), and MCP server integration. Use when asked to "set up copilot", "create an agent", "add instructions", "create a prompt", "scaffold a skill", "configure hooks", "customize copilot", or when building any GitHub Copilot customization for VS Code or GitHub.com.
What this skill does
# GitHub Copilot Customization Skill
Create and manage the full spectrum of GitHub Copilot customizations for VS Code (1.109+),
GitHub Copilot CLI, and Copilot coding agent.
## Decision Tree — Which Primitive to Use
Before creating anything, determine the right customization type:
```
User request
│
├─ "Set up Copilot for this project" or "/init"
│ → Generate .github/copilot-instructions.md (always-on instructions)
│ → See: references/instructions.md
│
├─ "Rules for specific file types" (e.g., "all .tsx files should...")
│ → Create .instructions.md with applyTo glob pattern
│ → See: references/instructions.md
│
├─ "Reusable task" (e.g., "scaffold a component", "generate tests")
│ → Create .prompt.md in .github/prompts/
│ → See: references/prompt-files.md
│
├─ "Specialized persona" (e.g., "code reviewer", "security auditor", "planner")
│ → Create .agent.md in .github/agents/
│ → See: references/custom-agents.md
│
├─ "Teach Copilot a complex capability with scripts/resources"
│ → Create skill folder in .github/skills/<name>/SKILL.md
│ → See: references/agent-skills.md
│
├─ "Run checks before/after agent actions" (lint, security, logging)
│ → Create hooks.json in .github/hooks/
│ → See: references/hooks.md
│
├─ "Connect external services" (databases, APIs, issue trackers)
│ → Configure MCP servers in .vscode/mcp.json or settings
│ → See: references/mcp-servers.md
│
└─ "Full project setup" or "enterprise customization"
→ Combine multiple primitives (see examples/enterprise-setup.md)
```
## Quick Reference: File Locations
| Primitive | Workspace Location | User/Profile Location |
|----------------|------------------------------------------|-------------------------------|
| Instructions | `.github/copilot-instructions.md` | VS Code profile folder |
| | `.github/instructions/*.instructions.md` | |
| Prompts | `.github/prompts/*.prompt.md` | VS Code profile folder |
| Agents | `.github/agents/*.agent.md` | VS Code profile folder |
| Skills | `.github/skills/<name>/SKILL.md` | `~/.copilot/skills/<name>/` |
| Hooks | `.github/hooks/*.json` | N/A (repo-level only) |
| MCP Servers | `.vscode/mcp.json` | VS Code settings |
## Quick Reference: Frontmatter Fields
### Instructions (.instructions.md)
```yaml
---
applyTo: '**/*.ts' # Glob pattern — when to apply (omit for always-on)
---
```
### Prompt Files (.prompt.md)
```yaml
---
description: 'Short description shown in / menu'
agent: 'agent' # Target agent: 'agent', 'ask', 'edit', or custom name
model: 'Claude Sonnet 4' # Optional model override
tools: ['search/codebase', 'githubRepo', 'read', 'edit'] # Available tools
---
```
### Custom Agents (.agent.md)
```yaml
---
name: my-agent # Display name
description: 'What this agent does and when to use it'
tools: ['search', 'read', 'edit', 'fetch', 'usages']
model: 'Claude Sonnet 4' # Optional model
handoffs: # Optional workflow transitions
- label: 'Next Step'
agent: other-agent
prompt: 'Continue with...'
send: false # true = auto-submit
---
```
### Agent Skills (SKILL.md)
```yaml
---
name: my-skill-name # Lowercase, hyphens for spaces
description: >
What the skill does and when Copilot should load it.
Be specific about trigger phrases and use cases.
---
```
### Hooks (hooks.json)
```json
{
"hooks": [
{
"type": "command",
"event": "pre-tool-execution",
"bash": "./scripts/my-hook.sh",
"powershell": "./scripts/my-hook.ps1",
"timeoutSec": 30
}
]
}
```
## Workflow: Creating Customizations
1. **Identify the need** — Use the decision tree above
2. **Read the reference** — Open the relevant file from `references/`
3. **Copy the template** — Start from the matching file in `templates/`
4. **Customize** — Follow the guidelines in the reference document
5. **Test** — Use the Chat Debug View (Ctrl+Shift+P → "Chat: Open Debug View")
6. **Iterate** — Refine based on actual agent behavior
## Progressive Discovery
This skill uses **progressive discovery** — only load what you need:
- **This file (SKILL.md)**: Decision tree + quick reference (always loaded first)
- **references/**: Deep-dive documentation per primitive (load on demand)
- **templates/**: Copy-paste starting points (load when scaffolding)
- **examples/**: Real-world configurations (load for inspiration)
When Copilot loads this skill, start here. Only read reference files when you need
detailed guidance on a specific primitive. Only read templates when you need to
generate files. Only read examples when the user wants a comprehensive setup.
## Important Notes
- Agent Skills require the `chat.useAgentSkills` setting enabled (preview feature)
- Hooks work with Copilot coding agent and Copilot CLI (not local chat)
- Custom agents were previously called "chat modes" — `.chatmode.md` files still work
- Skills in `.claude/skills/` directories are also recognized by Copilot
- The `/init` command auto-generates initial instructions from your project
- Use the Chat Debug View to verify which instructions/skills are loaded
Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.