Introducing Flows: Your Team's Microservice Command Center

Yiren Lu -

Why We Built Flows

While Plural has always provided powerful tools for platform engineers to manage
infrastructure-wide deployments, we recognize that application developers have
different needs. They typically focus on a smaller set of services and need a
more streamlined way to monitor, troubleshoot, and manage their microservices.

Enter Flows: a new developer-centric interface that brings together everything
you need to manage your microservices in one place.

What Is a Flow?

A Flow is a logical container that groups related microservices and their
development pipeline. Think of it as your microservice's mission control center,
where you can:

  • Monitor connected services
  • View deployment pipelines
  • Track related pull requests
  • Handle alerts
  • Access an AI-powered chat interface for troubleshooting

Core Features

Service Management and Pipelines

At its heart, a Flow helps you manage related services together. You can monitor
service health, track deployment pipelines, see related PRs and understand how services interact with each other. The pipeline view shows you exactly how updates flow from one
service to another, while integrated alerts from PagerDuty, Grafana, or other
tools keep you informed of any issues.

Built-in Governance

One of Flow's most valuable features is its permission boundary system. Instead
of managing permissions for each component separately, you can set them once at
the Flow level. This makes it easy to say "this team owns these services" and
ensure everyone has the access they need—and only the access they need. These
boundaries also help the AI assistant (see below) respect data privacy, ensuring it only
searches within the context you've defined.

Operational Insights

Each Flow gives you a clear window into your services' operations. You can view
aggregated logs, check performance metrics with heat maps, and track historical
data—all with the proper access controls in place. When something goes wrong,
you have all the context you need to understand and fix the issue.

AI-Powered Troubleshooting

Flows comes with Flow Chat, our intelligent AI-powered assistant for understanding and
fixing issues. It can help you explore your infrastructure state, explain
service configurations, and remediate errors.

Most relevantly, all the information that goes into the context of FlowChat comes from within the boundary of the Flow. In other words, when you ask it a question, it searches only the related PRs, logs, services. This helps improve the accuracy and actionability of its responses.

MCP Integration for Advanced Operations

Flows integrates with third-party MCP servers, which opens up
powerful possibilities for automation and tooling. You can connect to popular MCP
servers to handle common operational tasks without building custom admin
interfaces. We've seen teams use this to manage everything from database
operations to Slack notifications.

Moreover, the MCP integration comes with enterprise-grade controls built in. Every
operation is logged for audit purposes, permissions are granular, and sensitive
operations can require explicit confirmation. When an operation needs approval,
you'll see a preview of exactly what will happen before you confirm.

Automated Troubleshooting

In addition to the Flow Chat interface in the UI, Flows also helps improve Plural AI behind the scenes.

For example, in the microservice world, developers continuously ship code changes that can introduce regressions—500 error rates spike, latency increases, and other indicators signal potential instability.

Flows addresses this challenge with automated troubleshooting designed specifically for application code deployments. When alerts fire, the system automatically:

  • Analyzes alerts and associated logs
  • Cross-references recent pull requests from GitHub or GitLab
  • Identifies the exact code changes causing issues
  • Pinpoints specific commit SHAs and lines of code
  • Generates targeted fix recommendations

All of this happens within strict permission boundaries. Teams only see code they're authorized to access, preventing accidental exposure across departments or projects.

For example, the system can detect when new error-handling logic is causing issues at specific endpoints, trace it back to the responsible PR, and suggest remediation—all without manual intervention. This transformation of troubleshooting from reactive fire-fighting to proactive problem-solving dramatically reduces mean time to resolution (MTTR) and helps teams maintain service stability.

Getting Started

Setting up a Flow is straightforward. Flows are configured using Kubernetes
Custom Resource Definitions (CRDs). You'll define your Flow with a description
and optional icon, set up your permissions, and register the services you want
to include. Once that's done, you can enhance your Flow by connecting MCP
servers for additional automation capabilities.

For the full setup instructions, see here

Try Flows Today

Ready to simplify how your team manages microservices?

Book a demo to see Flows in action, or sign up to get started with Plural.

For technical details and setup instructions, check out our
documentation.