Flows v1.0
1. State Machine Engine
At the core of Flows is a state machine engine, allowing you to design and execute workflows. With this engine, you can manage processes efficiently, control transitions between states, and ensure every part of the workflow behaves as expected.
2. Triggers
Flows supports a variety of triggers to initiate workflows. Whether you're integrating external applications or internal systems, these triggers ensure your workflows respond dynamically:
Catch Hook – Activate workflows by listening for incoming webhooks from any source.
New Inbound Email – Start a workflow when specific emails are received.
Scheduler for Flows - Activate workflows at specified intervals. This includes:
Daily scheduling – Automatically trigger workflows at a set time each day.
Hourly scheduling – Execute workflows on an hourly basis.
The Scheduler makes it easy to implement time-based automation.
User modified – Trigger workflows when a user's information is updated.
Tenant modified – Respond to changes made to a tenant in Visma Connect IdP.
Self-service extend own Triggers
Developer Portal will allow you to extend Flows with your own Triggers. Both Visma apps and Third-party apps.
3. Actions
Flows enables users to define powerful actions, providing a wide range of integration and customization options:
HTTP Action – Make external HTTP requests as part of your workflow to any target.
Send outbound Email – Send custom emails directly from your workflows.
Javascript – Include custom JavaScript logic.
Python – Add Python scripts for advanced customization.
Add user – Seamlessly add users to Visma Connect IdP.
Self-service extend own Actions
Developer Portal will allow you to extend Flows with your own Actions. Both Visma apps and Third-party apps.
Functions
Flows also provides built-in function actions to handle various data processing tasks, including:
DateTime: Perform date/time comparisons and operations.
Double: Check equality between two doubles, or perform mathematical operations on them.
Long: Check equality between two integers, or perform mathematical operations on them.
Text:
Convert text from Base64.
Get the length of a text string.
Count the number of words in a text.
And many more to extend your workflow’s capabilities!
4. Retry and Redrive
Ensure workflow reliability with robust failure-handling capabilities:
Retry an entire execution to resolve temporary issues.
Redrive from a specific point of failure, minimizing downtime and reprocessing effort.
5. Looping
Flows offers native support for looping constructs, allowing workflows to handle repetitive tasks dynamically and efficiently.
6. If/Else Conditions (Branches)
Make your workflows smarter with built-in if/else conditions, enabling decision-based branching and greater control over workflow behavior.
7. Version Management
Flows makes it simple to manage and iterate on workflows with a robust versioning system:
Start with Version 1 of your flow, design it, and publish it.
When further changes are needed, a draft version is automatically created upon edit.
Continue modifying the draft version until it has been tested thoroughly and is ready for release.
Publish the draft as Version 2, while maintaining a clear history of previous versions for reference.
This version management system ensures safe updates and allows for iterative improvements to workflows without disrupting the currently active flow.
8. Execution Overview
Gain full visibility into your workflows in the execution overview, designed to keep you informed:
Monitor workflow statuses at a glance to quickly identify issues or track successful executions.
9. Workflow Overview
Overview of all your workflows:
Filter workflows based on applications, creators, dates.
Developer Portal admin for Flows
The Developer Portal serves as a gateway for developers to expand the capabilities of Flows by integrating their applications. Any Visma application created in the Developer Portal that leverages APIs or Webhooks can be published to Flows and showcased as a product.
By publishing your application in Flows, you can:
Define Triggers and Actions: Expose your application’s features as triggers and actions that users can leverage to build powerful integrations.
Enable Seamless Integration: Allow users to connect and automate processes with your application effortlessly.
Expand Product Reach: Position your application as part of the Flows ecosystem, making it accessible to a wider audience of developers and businesses.
The Developer Portal simplifies the process of turning your application into a valuable component of the Flows ecosystem, unlocking endless possibilities for automation and innovation.
Flows as an MCP Server for AI-Driven Automation
What is MCP (Model Context Protocol)?
The Model Context Protocol (MCP) (https://modelcontextprotocol.io/) is designed to enable seamless interaction between AI models and external systems. Through MCP, AI agents can:
Discover and interact with tools exposed by external products.
Execute actions such as sending messages, updating records, or retrieving data.
Dynamically integrate with different products and services without pre-programmed integrations.
MCP follows a host-client-server architecture:
Hosts: AI models (e.g., language models or other agents) needing access to external tools.
Clients: Middleware that facilitates communication between AI models (hosts) and systems (servers).
Servers: External systems (like Flows) that expose APIs for AI models to use.
Flows as an MCP Server
Flows will function as an MCP server, enabling each user to choose which products (tools) they want to expose to AI models for automation. This flexibility allows user to toggle any product with a chosen set of actions they have enabled inside Flows on/or off as a tool for AI agents.
Key points:
Users Choose Enabled Tools: Each user has the ability to enable or disable specific integrations directly from Flows' user interface. This includes Visma products and third-party products that behind the scenes are interfaced through APIs (e.g., Slack, CRM systems, ERP systems). If a tenant has a product enabled within their Flows environment, they can simply toggle it on as a tool for use with AI models.
Secure Credential Management: All credentials required for integrations (such as API keys, tokens, and authentication details for services like Visma) will remain securely stored inside Flows. This ensures that sensitive information is never exposed to or shared with external AI models (LLMs). Credentials are handled exclusively by Flows, protecting against potential security risks related to data exposure.
An advantage of using Flows as an MCP server is its ability to expose products that do not natively have their own MCP server for their API. This capability allows Flows to serve as an intermediary, enabling AI models to interact with a broader range of tools, whether they are Visma products or third-party services even if those tools do not have built-in MCP support. By keeping credentials securely stored within Flows, the system ensures that LLM clients never have access to the actual product credentials.