> For the complete documentation index, see [llms.txt](https://docs.wem.io/platform/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.wem.io/platform/release-notes/modeler-release/modeler-release-5.2.md). # Modeler Release 5.2 ### Highlights * **New "Invoke agent" node.** A new flowchart node that can trigger an AI agent without first sending a chat message, a complement to the existing "Send message" node, now also supporting file and image attachments. * **New agent selector interface.** Choose between multiple providers and their models, using the new model selector overlay for creating AI agents, with support for new tools like web search and code interpreters. * **Richer AI agent chatbox.** The chatbox component now supports a read-only mode and can optionally show tool calls, instructions, reasoning, and timestamps. ### New features #### AI Agents * **Invoke agent node;** New node that triggers an AI agent, independent of the "Send message" node. * **Editable provider field;** the provider field on existing AI agents can now be changed (previously only settable at creation). * **Deployment name syncing;** selecting an Azure OpenAI model now automatically syncs the deployment name to the model name (manual overrides remain possible). * **Reasoning effort;** a reasoning effort level can now be configured for supported AI models. * **Conversation dossier item type;** a new dossier type specifically for AI agent conversations (`AiAgentConversation`). * **New conversation functions;** `DateCreated()`, `DateModified()`, `AiAgentName()`, and `HasContent()` for the Conversation dossier item. * **Read-only chatbox mode;** the AI agent chatbox component can now be set to read-only, so it displays a live conversation without allowing input. * **Richer chatbox display options;** the chatbox can now optionally show tool calls (with their request/response messages), system/instruction messages, reasoning, and message timestamps, plus a customizable placeholder text for the prompt input field. * **Additional AI provider support;** added support for more AI providers, including Anthropic and Google Gemini, alongside OpenAI and Azure OpenAI. * **AI agent selector overlay;** creating a new AI agent now starts with a model selector overlay, showing provider, release date, reasoning capability, supported tools, and supported input types before you pick a model. * **File and image attachments;** the Invoke agent node now supports sending file and image attachments to the AI agent. * **Wait for response option;** the Send message and Invoke agent nodes both support choosing whether to wait for the agent's response or continue asynchronously. * **Web search and code interpreter support;** hosted tools like web search and code interpreter can now be enabled directly in the modeler, for AI models that support them. #### Functions * `ToConcept(localName, rootConcept)`**;** new overload of the `ToConcept` function. #### General * **Download public X509 certificate as PEM** instead of binary DER format. * **Documentation link added** to the CSP Automatic Switch overlay. * **Default icon** for the "Home" menu item on new projects. ### Bug fixes * Fixed a validation bug that prevented importing SAML metadata files. * Fixed autocomplete suggestions not working for authentication providers. * Fixed the 'Execute process' node's options not being visible. * Fixed the NavigationItemEditorOverlay incorrectly saving when canceling with an expression on the cancel button. * Fixed the DossierFieldSelectorOverlay incorrectly showing concept queries when adding an assignment node. * Fixed several OAuth2/JWT-related overlay issues: an error when opening the standard claim overlay, a missing info icon on custom claims, and an overly restrictive type validation on custom claim expressions. * Fixed a server validation error being shown incorrectly when saving a task with certain values. * Fixed flowchart node selection incorrectly persisting after cut/paste into another flowchart. * Fixed subdomain fields in the "Edit hostnames" overlay becoming misaligned with long hostnames. * Fixed the publisher overlay incorrectly showing a license warning when a portal already had a license attached. * Fixed expression properties in the Template Editor sometimes getting an unintended height of 13px. * Fixed form values in a node not being saved if the input field hadn't lost focus first. * Fixed an issue where importing a SOAP zip file with a custom XSD could throw an invalid URI error * Fixed an issue where a developer user couldn't open widgets from their personal workspace in the widget editor --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://docs.wem.io/platform/release-notes/modeler-release/modeler-release-5.2.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.