03 MAR 2026 - Fluxnova AI Working Group Meeting Minutes #18
Description
Date
03 MAR 2026 - 11am EST
Meeting Details
- Meeting link : https://zoom-lfx.platform.linuxfoundation.org/meeting/94324343329?password=4fd47917-b89f-4c7c-8925-a8a4e5b54ab1
- Register for future meetings: https://zoom-lfx.platform.linuxfoundation.org/meeting/94324343329?password=4fd47917-b89f-4c7c-8925-a8a4e5b54ab1&invite=true
Meeting notices
-
FINOS Project leads are responsible for observing the FINOS guidelines for running project meetings. Project maintainers can find additional resources in the FINOS Maintainers Cheatsheet.
-
All participants in FINOS project meetings are subject to the LF Antitrust Policy, the FINOS Community Code of Conduct and all other FINOS policies.
-
FINOS meetings involve participation by industry competitors, and it is the intention of FINOS and the Linux Foundation to conduct all of its activities in accordance with applicable antitrust and competition laws. It is therefore extremely important that attendees adhere to meeting agendas, and be aware of, and not participate in, any activities that are prohibited under applicable US state, federal or foreign antitrust and competition laws. Please contact legal@finos.org with any questions.
-
FINOS project meetings may be recorded for use solely by the FINOS team for administration purposes. In very limited instances, and with explicit approval, recordings may be made more widely available.
Agenda
- Convene & roll call (5mins)
- Display FINOS Antitrust Policy summary slide
- Review Meeting Notices (see above)
- Approve past meeting minutes - 24 FEB 2026 - Fluxnova AI Working Group Meeting Minutes #16
- Fluxnova BPM Platform MCP Monitoring Capability - Proposal by ScottLogic.
- Continue discussion Agentic AI Implementation analysis and next steps. #17
- MCP PR Follow-Up - ISSUE-10 Initial implementation for MCP Start Events. Includes a plug… #14
- OpenAI Connector Draft Feedback - Fluxnova OpenAI Connector #15
- AOB, Q&A & Adjourn (5mins)
Minutes
The meeting focused on discussing the integration of AI capabilities with Fluxnova, particularly through plugins and extensions. A demonstration from a recent industry hackathon showcased an MCP (Model Context Protocol) plugin for connecting Fluxnova with LLMs, which sparked a broader discussion about the architectural approach to AI integration.
The group reached a consensus to maintain AI features as separate plugins rather than integrating them into the core product. This "plugin-first" strategy allows for maximum flexibility and community customization while keeping the core engine stable. The team agreed to create a new dedicated repository for plugins, and a proposed directory structure will be documented shortly.
Next Steps
- Karl: Open a thread with help at FINOS to document and coordinate the creation of the
fluxnova-pluginsrepository. - Harish: Create an initial PR with the proposed directory structure and layout for the Fluxnova plugins repository once it is created.
- Pieter: Review the OpenAI connector draft feedback provided by Gabor this week.
- Chris: Look into enabling ad hoc sub process in Fluxnova core engine this week.
- All relevant participants: Review Pieter's analysis and ranking of Agentic AI implementation proposals, provide comments/feedback, and prepare for decision-making (to be discussed in maintainer's call on Thursday).
- Harish: Review Pieter's example in the Fluxnova Examples repository today or tomorrow.
- Jamie/Hackathon team: Continue development of the MCP plugin/Agentic AI integration over the next couple of weeks, focusing on security and determining where it fits in the plugin ecosystem.
- Pieter: Bring up the Agentic AI implementation analysis and decision for discussion at the maintainer's call on Thursday.
Detailed Summary
Fluxnova BPM Monitoring Proposal
The meeting opened with introductions and the approval of previous minutes. The chair welcomed new participants from a partner consultancy who were involved in a recent hackathon with Fluxnova. The main topic for discussion was the Fluxnova BPM platform MCP monitoring capability proposal, aimed at enhancing how users interact with and monitor running workflows through AI-driven tools.
AI-Driven Workflow Monitoring Demo
A project from a recent financial services AI hackathon was presented, featuring an MCP plugin for Fluxnova.
- Functionality: Provides agentic access to monitoring APIs, enabling LLMs to answer natural language queries about active trade risk calculations and workflow states.
- Visualization: The team discussed adding the ability to display BPMN process models as images within the chat interface using existing XML and BPMN JS rendering tools.
- Strategic Feedback: Maintainers raised points regarding the balance between AI automation and critical business visibility. It was suggested that milestone capabilities and custom reporting should remain the primary focus for high-stakes workflow transparency.
AI Integration and Plugin Strategy
The team finalized the strategy for integrating AI tools with Fluxnova:
- Plugin-Based Approach: Core orchestration and automation will remain separate from AI capabilities to ensure system modularity.
- New Repository: A separate repository will be established for plugins to encourage community contribution while maintaining strict control over the core functionality.
- Governance: The group noted the revamping of the FINOS Applied Gen AI working group, which will focus on monitoring hackathon projects post-event.
- Decision Pipeline: A detailed analysis of Agentic AI implementation proposals is currently under review, with a formal decision expected during the next maintainer's call.