AI Development Tools & Configurations

Practical configurations and working examples for AI coding assistants: multi-agent topologies, installable skills, and harness setup for Claude Code, OpenCode, and VS Code with GitHub Copilot.

This site complements the vendors' own documentation with ready-to-use configurations, integration patterns, and behavioral guidelines. It does not rewrite their guides. Learn more on the About page.

What do you want to do?

• Install a skill. Browse the skills catalog and follow its install commands.
• Install a topology. Choose between the Lane and Matrix multi-agent topologies, then follow the installation steps.
• Configure Claude Code. Start with the Claude Code overview: skills, the Matrix mirror, and Google Cloud Vertex AI setup.
• Configure OpenCode. Start with the OpenCode overview and the Configuration Guide.
• Configure VS Code with GitHub Copilot. Start with the VS Code guide and its Quick Start.
• Add an MCP server. See the MCP server examples.
• Contribute. Read the Contributing guide.

What's Inside

• Agents: two multi-agent topologies, Lane and Matrix, plus a roster of domain specialist agents (API, architecture, security, testing, and more).
• Skills: structured instruction sets that teach AI agents domain-specific tasks.
• Harness: configurations and guides for Claude Code, OpenCode, and VS Code.
• MCP Servers: working examples for Model Context Protocol server integrations.

Official documentation

• Claude Code
• GitHub Copilot
• Model Context Protocol
• OpenCode
• VS Code

Note

If you are an AI agent working in this repository, start with AGENTS.md. The pages on this site are publication content, not instructions.