Skip to main content

Backstage Agent

  • 🤖 Backstage Agent is an LLM-powered agent built using the LangGraph ReAct Agent workflow and Backstage MCP Server.
  • 🌐 Protocol Support: Compatible with A2A protocol for integration with external user clients.
  • 🛡️ Secure by Design: Enforces Backstage API token-based RBAC and supports secondary external authentication for strong access control.
  • 🏭 MCP Server: The MCP server is generated by our first-party openapi-mcp-codegen utility, ensuring version/API compatibility and software supply chain integrity.
  • 🔌 MCP Tools: Uses langchain-mcp-adapters to glue the tools from Backstage MCP server to LangGraph ReAct Agent Graph.

🏗️ Architecture​

System Diagram​

Sequence Diagram​

⚙️ Local Development Setup​

Use this setup to test the agent against a local Backstage instance.

▶️ Start Backstage with Docker​

Note: Backstage can be run locally using Docker for development and testing.

# Clone Backstage
git clone https://github.com/backstage/backstage.git
cd backstage

# Start Backstage with Docker Compose
docker-compose up -d

🛂 Retrieve Admin Credentials​

# Get the admin user token from Backstage logs
docker-compose logs backstage | grep "admin user token"

📦 Install CLI (Optional)​

# Install Backstage CLI
npm install -g @backstage/cli

🚀 Deploy Example Service​

# Create a new service using Backstage CLI
npx @backstage/cli create
# Follow the prompts to create a service

🔑 Get API Token​

  1. Log in to your Backstage instance
  2. Go to Settings → API Access
  3. Create a new API token with appropriate permissions
  4. Save the token for your .env file

Add to your .env:

BACKSTAGE_API_KEY=<your_token>
BACKSTAGE_API_URL=http://localhost:7007

Local Development​

# Navigate to the Backstage agent directory
cd ai_platform_engineering/agents/backstage

# Run the MCP server in stdio mode
make run-a2a