🚀 Platform teams can now query drift, compliance, and cloud resources directly via LLMs without leaving their workflow. StackGuardian's Model Context Protocol (MCP) brings governance data into AI tools like Claude or ChatGPT.
The Context Switching Problem
Platform engineers manage thousands of resources across multiple clouds, but constant tool hopping kills productivity. Manual dashboard navigation for drift checks or report exports delays critical decisions.
MCP eliminates this by exposing StackGuardian data through a standardized protocol for LLMs. Teams query in natural language, like "Show drifted resources in production," and get instant analysis.
What is MCP?
MCP is a protocol that lets LLMs access StackGuardian's drift detection, compliance, and governance data via a server endpoint. It supports tools like MCP servers for seamless integration into AI workflows.
Key capabilities include real-time queries across 100+ accounts and policy violation troubleshooting. No custom scripting or API wrangling required.
Technical Implementation
Deploy an MCP server to bridge StackGuardian and your LLM. It handles authentication, queries drift status, compliance reports, and resource inventories.
Integrate with any LLM supporting MCP, pulling live data without exports. Scales to enterprise cloud estates.
How It Works in Practice
MCP transforms everyday infrastructure challenges into simple conversations. Here's how platform engineers are using it:
- Debugging failed deployments
The problem: A production deployment failed overnight and you need to understand why.
The prompt: "Why did workflow run #1247 fail? Help me figure out what happened."
How MCP helps: The LLM retrieves execution logs, identifies the failure point (expired credentials, policy violations), and presents a clear explanation with relevant error messages. - Monitoring deployment status
The problem: You need to check if your latest changes deployed successfully without navigating through multiple dashboards.
The prompt: "What's the status of my latest workflow run?"
How MCP helps: The LLM instantly retrieves your most recent workflow execution and reports status, timestamps, and any issues. - Organization-wide infrastructure visibility
The problem: Your team manages dozens of stacks across environments. Getting a comprehensive health view requires checking each stack manually.
The prompt: "What's the current status of all stacks in my organization?"
How MCP helps: The LLM aggregates all stack statuses, identifies failures or pending changes, and presents a complete health summary. - Template version management
The problem: You need to identify which workflows are using outdated template versions before planning an upgrade rollout.
The prompt: "Which workflows are still using template empty-tf-resource revision 2?"
How MCP helps: The LLM scans all workflows, identifies those using the specified template revision, and provides a list with workflow names and configurations. - Workflow health assessment
The problem: Before a major release, you want to ensure all workflows in a group are healthy and ready for production traffic.
The prompt: "How healthy are the workflows in the mcp-testing workflow group?"
How MCP helps: The LLM analyzes all workflows in the group, checks recent run success rates, identifies recurring failures, and reports overall health metrics.
Why MCP Matters
MCP fuses infrastructure orchestration with AI workflows, standardizing governance access. Platform teams gain speed without sacrificing controls.
It's open for custom extensions, much like extensible IaC platforms.
Getting Started
- Set up your MCP server following StackGuardian docs.
- Configure your LLM (e.g., Claude) with the MCP endpoint.
- Query away: Start with drift or compliance checks.
Available now for StackGuardian users. Dive into docs at
https://docs.stackguardian.io/docs/mcp-server/
What's Next
We're expanding MCP for more policy-as-code integrations and multi-cloud depth. Join the conversation on our community channels.
Unlock AI-driven governance? Try MCP today →
