Apache 2.0 · Open Source v0.8.1

OpenShift Integration Operator

The missing lifecycle layer for Apache Camel on OpenShift

Design visually, deploy with GitOps, observe in real time — Apache Camel + CNCF SonataFlow, visual Kaoto designer, MCP/AI integration, and native Kubernetes orchestration.

OperatorHub Artifact Hub Vulnerability report on Artifact Hub Security dashboard on GitHub Pages OpenSSF Best Practices

Try it now — no prerequisites View on GitHub

Security report · Compare platforms · Contribute a Flow · Migrating from Camel K

The Evolution of Apache Camel

From a routing engine to the backbone of modern integration.

Born in 2007, Apache Camel evolved from a Java library implementing EIPs into a complete cloud-native ecosystem with Camel K and Camel Quarkus.

This operator takes all that historical power and wraps it in a modern, GitOps-native, and visual user experience.

Leer más →

Test any Camel route in 30 seconds

No Git, no pipeline, no friction — ephemeral Quick Try with TTL, automatic ownerReference cleanup, and one-click promote to GitOps.

Quick Try Mode

Set deploymentMode: EPHEMERAL and paste kaotoDesign — the operator deploys a precompiled worker immediately.

⏱

TTL & cleanup

Configurable TTL with extend in the console. Delete the CR — no orphaned Deployments thanks to built-in ownerReferences.

🔀

Promote to GitOps

When ready, promote from the console — scaffold Git repo, Tekton build, Argo CD sync.

Start the 10-minute eval track →

Who is this for?

A lifecycle and UX layer on top of Apache Camel — not a replacement runtime.

🛠

Platform engineer

CRD reconciliation, Tekton + Argo CD GitOps, multicluster ApplicationSets, OLM deployment from Quay.

Integration tech lead

Lift-and-shift from Camel K or Red Hat Fuse with better console UX and ephemeral try-before-GitOps.

Integration Flows list in OpenShift Console Visual saga flow diagram Kaoto designer embedded in console

See all 9 console screenshots ↓

250+ ready flows — add yours in 5 minutes

Contribute a template to the global catalog — every cluster picks it up via GitHub Pages. No operator upgrade required for catalog-only changes.

Contribute a Flow prepare-flow-contribution.sh

Features

Everything you need to design, deploy, and observe integration workflows on OpenShift.

Five Integration Types

Run Camel Routes, Kamelets, Pipes, Tests, or SonataFlow workflows from a single IntegrationFlow CR.

Quick Try (Ephemeral)

Deploy and test flows in minutes without Git, Tekton, or ArgoCD. Configurable TTL, extend, and promote to GitOps when ready.

Visual Designer

Embedded Kaoto canvas + SVG flow visualizer with branching logic and click-to-YAML navigation.

Template Catalog

Browse 250+ ready flows from the console create form — search, filter by category, and deploy in one click.

GitOps Native

Scaffold to Git → Tekton builds Quarkus workers → Argo CD syncs base/ manifests. Eight catalog examples in k8s/examples/.

MCP/AI Ready

Model Context Protocol bridge for discovering and invoking AI tools in flows.

Observable

OpenTelemetry + OTel Collector with real-time SSE telemetry for canvas node coloring.

Auto-scaling Workers

HPA v2 per IntegrationFlow — each worker Deployment scales independently on CPU and memory.

Advanced Console UI

Search, filter, paginate flows. Direct links to Pods, Tekton Pipelines, and ArgoCD apps.

Multi-cluster

Target multiple clusters from a single IntegrationFlow spec via ArgoCD ApplicationSets.

MCP AI Integration: The Future of Automation

Connecting Model Context Protocol with Apache Camel to create intelligent flows.

The Model Context Protocol (MCP) is an open standard that allows artificial intelligence models to access external tools and data securely. By integrating MCP with Apache Camel, we have created a revolutionary bidirectional bridge.

How does it work? AI models can now dynamically discover and invoke any Camel route as if it were a native tool. In turn, Camel flows can query AI models to make complex routing decisions, extract data from unstructured text, or generate personalized responses.

This transforms static integrations into intelligent agents capable of adapting to context in real-time, all while maintaining the robustness and scalability of Kubernetes.

Discover more about MCP AI →

MCP AI Flow Architecture

How it compares

Apache Camel K remains the runtime underneath — this operator adds lifecycle, GitOps, console UX, and multicluster targeting.

Criterion Camel K Red Hat Fuse MuleSoft This operator
Native GitOps (Tekton + Argo CD) Partial Limited No Yes
Visual designer in console No Fuse Online (EOL) Proprietary Kaoto + plugin
Multicluster ApplicationSet Manual No SaaS spec.targeting
Ephemeral try + promote No No No Yes
Open source / on-cluster Yes EOL No Yes (Apache 2.0)
Low-code iPaaS No No Different category Enterprise Camel lifecycle — not an iPaaS competitor

See the full platform comparison, Architecture, and Migrating from Camel K.

Quick Start

Evaluators: Track A — 10 minutes with ephemeral. Production: Track B — full GitOps stack.

# Install from Quay (OLM)
operator-sdk run bundle quay.io/maximilianopizarro/openshift-integration-operator-bundle:v0.8.1 \
  --namespace openshift-integration \
  --install-mode AllNamespaces \
  --timeout 10m

# First flow — ephemeral Quick Try
oc apply -f k8s/examples/09-ephemeral-demo.yaml
oc get integrationflow ephemeral-camel-demo -n openshift-integration -w

See the full Quick Start Guide for GitOps examples, console plugin access, and verification steps. Browse 250+ Ready Flows or the Examples Catalog (310 Apache Camel Quarkus components). See the Security Report.

Console Plugin

Integration Platform embedded in the OpenShift Console — manage flows, monitor services, and design workflows visually.

Integration Flows list page with search bar, integration type filter dropdown, phase filter badges, pagination controls, and direct resource links to Pods and Tekton Pipelines

1 — Browse flows

Integration Flows — start here; search, filters, links to Pods and Builds
Visual SVG flow diagram showing saga orchestration with branching logic, choice nodes, parallel multicast, and click-to-YAML node highlighting

2 — Inspect visually

Visual diagram — saga branching; click nodes for YAML
Kaoto drag-and-drop visual route designer embedded in OpenShift Console with component palette, canvas nodes, and property editor panel

3 — Edit in Kaoto

Kaoto designer — edit routes visually in the console
YAML editor showing kaotoDesign Camel route definition with syntax highlighting and automatic Git sync to configured repository

4 — Sync to Git

YAML editor — kaotoDesign with Git sync (GitOps mode)
IntegrationFlow spec and status panel showing deployment mode, engine type, targeting strategy, resilience settings, circuit breaker state, and lifecycle controls

5 — Configure spec

Spec & status — resilience, targeting, lifecycle controls
Platform Status dashboard with real-time health indicators for Operator API, Kaoto Designer, Gitea, ArgoCD, Tekton Pipelines, and OpenTelemetry Collector

6 — Check platform health

Platform status — Operator, Kaoto, Gitea, Argo CD, Tekton health
Flow Overview dashboard with KPI summary cards for total flows, running count, error rate, and distribution charts by integration type and deployment mode

7 — Monitor KPIs

Flow overview — KPIs, plugin version, catalog mode
Flow Logs tab with live worker pod log streaming, container selector dropdown, configurable tail lines, and follow mode toggle

8 — Stream logs

Flow logs — live worker pod streaming (ephemeral or GitOps)
Lifecycle Controls showing Pause, Resume, and Stop actions with confirmation modal dialog and current flow phase indicator

9 — Manage lifecycle

Lifecycle — pause, resume, stop with confirmation

Precompiled Worker Images

Domain-specific Camel Quarkus workers on Quay — auto-selected from your route, or pin with spec.ephemeral.workerImage.

Precompiled Worker Images Architecture 
# Default (all domains — 80+ extensions)
quay.io/maximilianopizarro/camel-worker-full:v0.8.1

# Domain-specific (smaller, faster startup)
quay.io/maximilianopizarro/camel-worker-core:v0.8.1        # timer, log, direct, seda, bean
quay.io/maximilianopizarro/camel-worker-messaging:v0.8.1   # kafka, amqp, jms, mqtt, nats
quay.io/maximilianopizarro/camel-worker-http:v0.8.1        # platform-http, rest, graphql, grpc
quay.io/maximilianopizarro/camel-worker-data:v0.8.1        # sql, jdbc, mongodb, file, ftp
quay.io/maximilianopizarro/camel-worker-cloud:v0.8.1       # aws2-s3/sqs/sns, azure, google
quay.io/maximilianopizarro/camel-worker-ai:v0.8.1          # langchain4j-chat, djl

Override auto-selection

spec:
  deploymentMode: EPHEMERAL
  ephemeral:
    ttlSeconds: 3600
    workerImage: quay.io/maximilianopizarro/camel-worker-messaging:v0.8.1
Full Worker — Included Extensions

Core & EIP: timer, log, direct, seda, bean, mock, quartz, cron, exec, controlbus, scheduler
Messaging: kafka, amqp, jms, activemq, mqtt5, nats, pulsar, stomp, sjms2, azure-servicebus, google-pubsub, knative
HTTP/API: platform-http, http, rest, rest-openapi, vertx-http, netty-http, graphql, grpc, websocket, coap
Data: sql, jdbc, mongodb, cassandra, couchdb, infinispan, redis, elasticsearch, solr, minio, file, ftp
Cloud: aws2-s3/sqs/sns/ddb/lambda/kinesis/ses/bedrock, azure-blob/cosmos/eventhubs, google-storage/bigquery/pubsub
AI/ML: langchain4j-chat, djl
SaaS: debezium-mysql/postgres/mongodb, salesforce, slack, telegram, mail, git, github, jira, ssh, hashicorp-vault
Healthcare/IoT: fhir, hl7, mllp, snmp
Transform: jackson, jsonpath, jaxb, csv, xslt, jolt, jslt, freemarker, mustache, velocity, flatpack, tika
Resilience: microprofile-fault-tolerance, microprofile-health, micrometer, opentelemetry

Pick the right worker → Worker Image Selector

Architecture

A Kubernetes-native platform that connects visual design, GitOps, and runtime observability.

The operator reconciles IntegrationFlow custom resources through scaffolding, Git push, Tekton builds, and ArgoCD deployment. The console plugin embeds Kaoto and streams live telemetry over SSE to color canvas nodes in real time.

Read Architecture Docs →