| TIBCO Flogo® (formerly known as TIBCO Flogo® Enterprise) enables AI orchestration and intelligent automation through a visual flow designer integrated into Visual Studio Code. Built on a high-performance Go (Golang) engine, it delivers superior runtime efficiency, an ultra-light memory footprint, fast startup, and low-latency processing across edge, cloud, serverless, and on-premises environments. With its AI-ready architecture and powerful set of connectors, Flogo® transforms enterprise data into intelligent, event-driven, AI-ready workflows. To understand all the capabilities, see Build Event-driven Apps with an Ultralight Framework. |
| Read the documentation |
What's new in version 2.26.5 (June 2026)
Upgrade Google Connection to Override Service Account Key: You can securely configure Google connections by leveraging enhanced options for the Service Account Key. You can override the key using a Base64-encoded value, a JSON service account key, a file path (starting with file://), or a Kubernetes secret, providing greater flexibility in managing your Google connections. For more information, see TIBCO Flogo® Connectors User Guide.
TIBCO Flogo® Connector for Agentic AI - Tech Preview:
Azure OpenAI LLM Provider support: A new AzureOpenAI provider option in the LLM Provider Configuration enables direct integration with Azure-hosted OpenAI models using Azure-specific deployment, region, and resource settings.
Memory Conversation Store in the LLM Client Activity: The LLM Client Activity now supports an in-process memory conversation store, enabling multi-turn conversational interactions.
Dynamic MCP Server and A2A Server configuration via LLM Client Activity inputs: The LLM Client Activity Input tab now includes mcpServerConfigs and a2aServerConfigs array fields, allowing MCP servers and A2A remote agents to be configured dynamically at runtime without pre-configured connection resources.
TIBCO Flogo® Connector for Model Context Protocol (MCP):
Scope-based access control for JWT authentication: A new Required Scope field on MCP Server handlers enables per-tool, per-prompt, and per-resource authorization based on JWT token scopes.
Structured content support for tool responses: A new Reply Settings tab with Tool Output Schema and a structuredContent field in Map from Flow Outputs enable MCP tools to return structured, schema-validated responses alongside text responses.
Content annotations for tool responses: The MCP Server trigger reply now includes optional audience and priority fields on the tool response object, enabling MCP content annotations on tool results. For existing apps, you must sync the trigger reply schema to pick up the new audience and priority fields before they can be mapped in the flow.
Enhanced tokenInfo claims: The tokenInfo object in Map to Flow Inputs now exposes additional standard JWT claims (iss, sub, aud, name, email) for richer authorization and audit logging.
MCP Client Activity: A new activity that enables Flogo flows to invoke tools on remote MCP servers as an MCP client, with support for multiple transport types, authentication modes, and structured JSON responses.
MCP Client OAuth2 Configuration: A new connection type that provides OAuth2 authentication for the MCP Client Activity, supporting both Client Credentials and Authorization Code grant types.
OpenTelemetry support for MCP Server trigger: The MCP Server trigger now supports OpenTelemetry for distributed tracing and observability of MCP server operations.
TIBCO Flogo® Connector for Apache Pulsar:
Trigger-level Flowlimit: TIBCO Flogo® Connector for Apache Pulsar now supports trigger-level flowlimit.
For complete release details, see TIBCO Flogo® 2.26.5 Release Notes.
When working with TIBCO Flogo®, you may find the following videos useful:
Building AI Agents in minutes: Agentic AI Orchestration for the Modern Enterprise using TIBCO Flogo® | Build Integration Flows in Minutes Using Natural Language Prompts | TIBCO Flogo® AI Design Assistant | How to Build an AI Alert Agent with TIBCO Platform & Flogo |
Introducing TIBCO FLOGO Model Context Protocol (MCP) Connector | Installing TIBCO Flogo Extension for VS Code and Creating first Flogo App | Creation of Unit Testing & Play Test Case in Flogo |
Create your first REST API | Flogo Quick Start 2.21.0 | Deploy to AWS Lambda |
Subflows | Build App Binary | Adding Flogo Extensions via Github URL |
Application metrics | Turning your Flogo app into a GraphQL Server implementation | App configuration with environment variables |
CI/CD + Devops demo |
Accelerate your development by importing and customizing our predefined samples. These templates demonstrate how to build, test, and deploy Flogo applications using advanced capabilities. Each sample is categorized and includes a comprehensive guide. Visit the Flogo GitHub Repository, follow the instructions in the README file, and import them directly into your workspace.
AI-powered agents with custom guardrails, multi-agent handoff, MCP server integration, A2A (Agent-to-Agent) protocol, LLM Client Activity, dynamic tool selection at scale, scheduled reasoning with report generation, and incident triage.
Integrate business data (customers, products, sales) with AI Agents using the TIBCO Flogo® MCP Connector for natural language querying.
Explore production-ready REST, GraphQL, and gRPC samples.
Implement pre-built Flogo connectors for databases, messaging systems, and ecosystem integrations.
Learn how to create robust unit tests, handle assertions, and mock activities.
Deploy and scale your Flogo applications seamlessly as Azure Functions.
Package and deploy your Flogo applications using tailored, custom container images.
Join more than 60,000 TIBCO experts!
TIBCO® Community is your place to be engaged with other TIBCO users around the globe and get your questions answered.
The TIBCO® blog - Tips, Tricks, and Industry Trends