SF Awesome Copilot

Experimental catalog of skills, agents and prompts for GitHub Copilot.

Inspired by github/awesome-copilot.

Note: This project is experimental and the content is largely AI-generated. Use with critical thinking.

Structure

├── agents/          # Custom Copilot agents
├── skills/          # Skills organized by technology

Skills

Skill	Description	Path
glab	GitLab CLI -- issues, merge requests, CI/CD pipelines, repositories	skills/system/glab/
gh	GitHub CLI -- issues, pull requests, Actions, releases, PR review comments	skills/system/gh/
playwright-cli	Browser automation -- web testing, screenshots, form filling, data extraction	skills/system/playwright-cli/
skill-creator	Create, iterate, and benchmark agent skills with eval-driven workflows	skills/system/skill-creator/
doc-coauthoring	Structured workflow for co-authoring docs, proposals, and technical specs	skills/system/doc-coauthoring/
create-agentsmd	Prompt for generating an AGENTS.md file for a repository	skills/system/create-agentsmd/
code-security-audit	Multi-stack code security audit -- automated scans (Docker-based) + manual deep review	skills/system/code-security-audit/
agentic-security-audit	AI agent security audit -- instruction files, MCP configs, LLM integration, OWASP Agentic Top 10	skills/system/agentic-security-audit/
githuman	Review AI-generated code before committing via GitHuman Docker instances (sjust/ajust)	skills/system/githuman/

Skills are documents that provide context to Copilot on specific topics. Each skill contains:

• Description and use cases
• Practical examples
• Commands and code snippets

Browse the skills/ directory for available categories.

Agents

Agents are custom Copilot configurations with specific instructions.

Browse the agents/ directory for available agents.

Usage

Prerequisites

⚠️ Warning: Agent Skills support in VS Code is currently in preview and only available in VS Code Insiders.

To enable Agent Skills:

1. Install VS Code Insiders
2. Enable the setting in your workspace or user settings:

Via Settings UI:

• Open Settings (Ctrl+, or Cmd+,)
• Search for chat.useAgentSkills
• Enable the checkbox

Via JSON configuration (.vscode/settings.json):

{
  "chat.useAgentSkills": true
}

Installing Skills

1. Clone this repository
2. Copy the skills you need to your project's .github/skills/ directory
3. Copilot will automatically load them when relevant

Installing Agents

1. Copy the agent files (.agent.md) to your project's .github/ directory
2. Reference them in Copilot chat using @agent-name

Configuring MCP Servers (Optional)

MCP (Model Context Protocol) servers extend Copilot's capabilities by providing access to external tools and documentation.

Context7 MCP Server (Recommended)

Context7 provides up-to-date, version-specific documentation for libraries and frameworks. Get a free API key at context7.com/dashboard.

Create .vscode/mcp.json in your workspace:

Remote server (recommended):

{
  "mcpServers": {
    "context7": {
      "url": "https://mcp.context7.com/mcp",
      "headers": {
        "CONTEXT7_API_KEY": "YOUR_API_KEY"
      }
    }
  }
}

Local server:

{
  "mcpServers": {
    "context7": {
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_API_KEY"]
    }
  }
}

Start the MCP server: Open Command Palette (Ctrl+Shift+P) and run MCP: List Servers

Add a rule to auto-invoke Context7 (optional):

• Go to Cursor Settings > Rules or create .github/copilot-instructions.md
• Add: Always use Context7 MCP when I need library/API documentation, code generation, or configuration steps.

Contributing

Contributions are welcome. Keep the style dry and practical.