This MCP server template demonstrates how to orchestrate a council of AI agents across a shared codebase with a GitHub Projects v2 work queue, automated task delegation, and security-focused governance. The containerized server runs inside Docker and is designed to coordinate multiple specialized MCP servers (such as code quality, content creation, and security tooling) to execute layered agent workflows. To use it, configure your .mcp.json to point to the desired servers, provide any required API keys via environment variables, and start the Docker container. The orchestration logic includes flows for issue creation, agent claims, PR reviews, and human merges, enabling automated yet audited AI-driven development cycles. You’ll be able to leverage the included agent tooling for tasks like code generation, code review, security hardening, and CI/CD orchestration, while keeping sandboxing and trust boundaries in place.

Prerequisites:

• Docker Engine (v20.10+) and Docker Compose (v2.0+)
• Git
• Access to required API keys (OpenRouter, Gemini, etc.) if you plan to enable AI features

Installation steps:

1. Clone the repository git clone https://github.com/AndrewAltimit/template-repo cd template-repo

2. Ensure Docker is running and pull the MCP server image (this is a template image in this repo) docker pull template-repo/mcp-server:latest

3. Run the MCP server container

   This uses a minimal run command defined in the mcp_config; adapt as needed for your environment

   docker run -it --rm
   -e OPENROUTER_API_KEY=your-openrouter-api-key
   -e GEMINI_API_KEY=your-gemini-api-key
   -e CODEX_ENABLED=false
   template-repo/mcp-server:latest

4. Verify the server is healthy and consult the MCP configuration documentation (docs/mcp/README.md) for configuration strategy, required vs optional settings, and how to customize MCP servers.

5. Optional: use Docker Compose if provided in the repo to manage multi-container setups and multi-server orchestration.

Prerequisites recap: ensure Docker is installed and running, obtain any API keys you intend to use, and have a local copy of the repository to reference the MCP configuration guidance.

Tips and notes:

• The template emphasizes a container-first architecture; all MCP tooling runs inside Docker for portability.
• OpenAI/Codex integrations are disabled by default for security, per the project’s security stance. You can re-enable at your own risk if appropriate for your environment.
• Environment variables like OPENROUTER_API_KEY and GEMINI_API_KEY are optional but recommended if you plan to use AI agents for code generation, review, or other tasks.
• The MCP configuration is typically managed via a .mcp.json or similar config placed at the repo root; consult docs/mcp/README.md for the exact strategy and what needs to be declared for each server.
• If you encounter issues with container permissions or network access, verify that Docker has the required privileges and that any API endpoints are reachable from the container network.
• For CI/CD and security tooling, ensure your shell environment is properly set up to supply necessary secrets, and consider using secret management (e.g., Docker secrets or environment variable vaults) in production deployments.