Detecting and Correcting BPMN Models with LLMs
A tool for analysing and improving BPMN process models with large language models. It combines a graph view of the process with LLM-driven detection of modelling issues and suggested corrections, wrapped in an interactive web interface.
What it does
Visualises BPMN models and parses them for analysis.
Detects issues and proposes improvements via an LLM, with the changes highlighted visually against the original.
Accepts both text and image input — a BPMN diagram can be described or uploaded as a picture and reconstructed.
Validates and corrects models in real time, with a workflow for the user to accept or reject each suggested change.
How it is built
Processing leans on pm4py for BPMN handling and
networkx for the underlying graph operations, with
langchain orchestrating the LLM calls. The front end is a
Streamlit application embedding a BPMN viewer. A command-line path also
exists: the analyser and the fix-applier can be run as modules to batch
results into a spreadsheet for evaluation.
Source
Code and usage notes are in the source repository.