Agent Commons

Content Type: Module
Categories: Artificial Intelligence

Overview

Integrate the possibility to visually define agents with this module. Combined with other assets like GenAI Commons, Conversational UI and the connectors for models of OpenAI, Amazon Bedrock and Mendix Cloud GenAI, you have all the things you need to build smart, agentic AI apps with Mendix.

 

This module contains a domain model, pages and toolbox operations to define and call agents. It allows developers and prompt engineers to collaborate and build agentic systems.

 

See the Agent Builder Starter App for a full app with a template support agent for an example of how to use this module.

 

Create agents grounded in private knowledge like static reference guides and that can take action with tools, to call APIs or to retrieve live data in the app, like tickets.

 

Explore the full potential of generative and agentic AI in your Mendix applications with this starter app template and start building today!

 

Contact Us

For support and questions, feel free to reach out via email genai-components-feedback@mendix.com or Community Slack.

Documentation

This module contains a domain model, pages and toolbox operations to define and call agents. It allows developers and prompt engineers to collaborate and build agentic systems.

View documentation on Mendix docs.

General Setup

  1. Download the latest versions of the following modules from the Marketplace and follow the setup instructions: GenAICommons, ConversationalUI. A connector that is compatible with GenAI Commons is required for the interactions with your LLMs of choice.
  2. In the project security of your app, add the module role 'AgentCommons.AgentAdmin' to the user roles that are intended to define and refine Agents.
  3. Add the 'Agent_Overview' page (USE_ME > Agent Builder) to your Navigation or add the 'Snippet_Agent_Overview' to a page that is already part of your Navigation. Now the Agents can be defined at runtime.

Define Agents

  1. Run the app (locally or in the cloud).
  2. Create and edit agents at runtime. Write the prompts, use variable placeholders, add microflows as tools and connect knowledge bases to define the tasks and cabapilities of the agents.
  3. Configure the deployed model (LLM) that the agent should use, and connect the agent to it.
  4. Create various versions of the agents and test in an isolated setup to rapidly iterate, compare and evaluate the agentic behavior
  5. Set a version as "in use" to allow it to be called from the actual logic in the app.

Call an Agent from App Logic

Use toolbox operations "Call Agent" and "Create Chat for Agent" to use an agent in a microflow and make it part of your app logic. Take a look at our documentation or download the Agent Builder Starter App from the Marketplace for a detailed example of how to use the "Call Agent" activity in an action microflow of a chat interface.

Releases

Version: 1.2.0
Framework Version: 10.21.0
Release Notes: Added an additional toolbox action for developer convenience. There are now two Call Agent toolbox actions in total: - Call Agent Without History (new): This simple action allows to invoke the agent with the user prompt defined in the agent version, without any prior conversation history. - Call Agent With History (improved): This action enables users to pass a conversation history along with the request, allowing for more context-aware interactions. - Both actions support optional context objects for variable replacements in prompts. UX improvements: - Added more informative text throughout the interface to guide users effectively. - Improved the selection process for tool microflows and knowledge base collections, making it easier to find the right resources. - Streamlined the editing of agent details during testing, allowing for real-time adjustments. Bug fixes: - Resolved an issue where optional parameters (temperature, topP, and max tokens) were not being applied correctly in certain cases when testing a single-call agent in the Agent Builder test UI. - When importing and exporting agents, optional parameters are now persisted as well.
Version: 1.1.0
Framework Version: 10.21.0
Release Notes: We have added a new action that allows you to call agents with a context object to replace placeholders in a prompt.
Version: 1.0.0
Framework Version: 10.21.0
Release Notes: Building blocks to create agents at design-time and run-time.