Data Flow
How are rug pulls computed?
The following figure describes the dataflow of both Kleio and the webapp used to display the results (kleio/models contains detailed documentation of its architecture, as well as neo4j and mongodb structures).
The extraction part of the flow is carried out by Kleio itself. The data extracted from the workflows, components (GHAs), and vulnerabilities is then saved in a Neo4j database.
The second part of the flow, namely the detector, is carried out by the scripts contained in this webapp. For convenience and replicability purposes, we already provide the CSV files produced by the scripts. Such files can be found in the data/directory. If you are cloning this repository from GitHub, check Dataset.
Finally everything contained in data/ is used as input by the webapp to display statistics and Gantt charts.