GitHub Copilot SuperPowers

Opinionated, repo-local configuration for GitHub Copilot (Chat + Agent mode): instruction hierarchy, reusable prompt files, custom agent profiles, and MCP server config.

Quickstart • What’s inside • MCP servers • Repo layout

[!NOTE] This repository is intentionally small and highly opinionated. It’s designed to be copied into other repos and adapted.

Why this exists

GitHub Copilot gets significantly more useful when you make its “defaults” explicit: what quality bars to follow, when to stop and investigate, and how to keep edits scoped. This repo packages those defaults in a form that’s easy to version, review, and reuse.

What’s inside

Instruction system under .github/instructions/ (rules, workflow, quality gates, execution protocol)
Reusable prompts under .github/prompts/ (ex: generate a README, generate AGENTS.md, repo story time, create release notes)
Custom agents under .github/agents/ (ex: Critical Thinking, Mentor, Janitor)
Agent "skills" playbooks under .github/skills/ (investigation mode, root-cause tracing, verification gates, uncertainty verification, minimalist development, task direction approval)
VS Code workspace settings under .vscode/settings.json (enables Copilot agent features, advanced reasoning, MCP integration)
MCP configuration under .vscode/mcp.json (Sequential Thinking, Context7, Memory, Serena)

Quickstart

Use this repo as a template

Copy these folders into your target repository:
- .github/
- .vscode/
Review and tailor the instruction files:
- Start with .github/copilot-instructions.md
- Then adjust rules in .github/instructions/ to match your team’s tolerance for “guard rails”.

[!IMPORTANT] The instruction set in .github/instructions/ is designed to reduce flakiness and “agent drift”, but it can also feel strict. If you don’t want delegation-heavy workflows or you’re not using subagents, loosen or remove those constraints.

Use it in VS Code

Open the repository (or your repo that copied these files) in VS Code.
Ensure GitHub Copilot is enabled.
Keep .vscode/settings.json checked in so the workspace defaults travel with the repo.

Agent Skills

Agent skills are optional playbooks that can be loaded on-demand to handle specific scenarios. Load a skill when you need specialized handling for complex tasks.

Located in .github/skills/, this repo includes 6 ready-to-use skills:

Skill	Purpose	When to use
investigation-mode	Pause implementation after 2+ consecutive failures and switch to root-cause analysis before resuming	Repeated test failures or validation errors
root-cause-tracing	Trace bugs backward through call stacks to find the original trigger, not just the symptom	Deep debugging of complex issues
task-direction-approval	Explain failure root cause, present 2-3 solution options with trade-offs, and wait for explicit user choice	Before switching tech stacks or architectures
uncertainty-verification	Verify exact commands, configuration keys, API details, and version-specific behavior via official docs—no assumptions	When providing version-dependent or time-sensitive guidance
verification-before-completion	Run verification commands and confirm output before making any completion claims	Before marking tasks as complete or fixed
minimalist-surgical-development	Prioritize the smallest diff first, prefer standard libraries, avoid unsolicited refactoring	Code changes and refactoring work

Custom Agents

Custom agents define specialized "modes" for focused collaboration. Select an agent profile when you want tailored behavior for your specific task.

Located in .github/agents/, this repo includes 4 custom agent profiles:

Agent	Role	Best for
Critical Thinking	Questions assumptions and encourages exploration of alternative approaches and edge cases	Architecture reviews, design decisions, risk analysis
Mentor	Provides Socratic guidance through critical questioning without providing direct solutions	Learning, skill development, understanding complex systems
Janitor	Focuses on cleanup, simplification, and technical debt remediation	Code cleanup, refactoring, removing dead code, performance optimization
Tech Debt Remediation Plan	Generates comprehensive technical debt remediation plans with metrics and actionable steps	Technical debt analysis, code quality assessment, planning refactoring work

MCP Servers

This repo includes a ready-to-use MCP configuration at .vscode/mcp.json.

Model Context Protocol (MCP) servers extend Copilot's capabilities with specialized tools and information sources.

Configured servers:

Server	Purpose	Launch method
Sequential Thinking	Structured multi-step reasoning with built-in revision support for complex problem-solving	`npx`
Context7	Up-to-date library and framework documentation lookup—always retrieves current version docs	`npx`
Memory	Lightweight knowledge graph for storing discovered patterns, architectural insights, and recurring solutions	`npx`
Serena	Project-aware code intelligence: symbol operations, file search, refactoring, and codebase analysis	`uvx` (Python `uv` tooling)

Prerequisites & Troubleshooting

Node.js: Required for npx-based servers (Sequential Thinking, Context7, Memory)
Python uv tooling: Required for Serena MCP server via uvx
No MCP servers starting? Verify Node.js is installed (npm --version) and uv is available (uv --version)

VS Code Settings

Located in .vscode/settings.json, the workspace includes optimized Copilot agent configuration.

Agent & Skill Activation

Setting	Purpose
`chat.agent.maxRequests`	Limits agent requests per session (set to 125)
`chat.useAgentSkills`	Enables loading skills from `.github/skills/` for specialized task handling
`chat.useAgentsMdFile`	Loads custom agent profiles from `.github/agents/*.agent.md`
`chat.useNestedAgentsMdFiles`	Supports multi-level agent profile organization
`chat.customAgentInSubagent.enabled`	Allows subagents to use custom agent profiles for complex delegation

Advanced Reasoning & Code Quality

Setting	Purpose
`github.copilot.chat.responsesApiReasoningEffort`	Set to `"high"` for deeper problem analysis and fewer mistakes
`github.copilot.chat.responsesApiReasoningSummary`	Set to `"detailed"` for thorough thinking explanations
`inlineChat.enableV2`	Enables inline chat suggestions during code editing
`github.copilot.nextEditSuggestions.enabled`	Provides contextual edit suggestions based on chat history

Context & Search

Setting	Purpose
`github.copilot.chat.newWorkspace.useContext7`	Automatically enables Context7 MCP for docs lookup in new workspaces
`github.copilot.chat.scopeSelection`	Limits context to the current selection for focused prompts
`github.copilot.chat.codesearch.enabled`	Enables semantic code search in chat context
`github.copilot.chat.alternateGptPrompt.enabled`	Provides alternative prompt rewriting suggestions

MCP & CLI Integration

Setting	Purpose
`github.copilot.chat.useResponsesApi`	Enables reasoning and extended analysis capabilities
`github.copilot.chat.cli.customAgents.enabled`	Allows custom agents to be used from the CLI
`github.copilot.chat.cli.mcp.enabled`	Enables MCP servers in CLI mode for agent workflows
`github.copilot.chat.githubMcpServer.enabled`	Activates GitHub MCP for repo operations (issues, PRs, code search)
`github.copilot.chat.githubMcpServer.readonly`	Restricts GitHub MCP to read-only mode (set to `true` for safety)

Analytics

Setting	Purpose
`editor.aiStats.enabled`	Tracks AI-assisted code writing metrics for insights

Repo layout

hljs

.
├─ .github/
│  ├─ copilot-instructions.md
│  ├─ agents/
│  ├─ instructions/
│  ├─ prompts/
│  └─ skills/
└─ .vscode/
   ├─ mcp.json
   └─ settings.json

Notes

[!CAUTION] GitHub-flavored admonitions (like the blocks in this README) don’t render everywhere and are intentionally limited in where they can be nested. If you copy/paste sections into places like <details> or nested lists, you may need to rewrite them as plain blockquotes.

github-copilot-superpowers