THALis: README and figure provenance
README figures need breadcrumbs back to the run. data id + config + script near the image would save so much hunting later
documentation commits: 4f4b43e adds the project title and status; 124da36 adds figures under img/. The archive identifies TDA plots, a collector screenshot and an EEG-streaming visualization.
Virtual-hand collector interface
reference use: the README is an entry point into the implementation, not the source of truth for every numeric result. Each displayed image should point back to its input data, analysis configuration and generating script.
review questions: which figure is a live acquisition screenshot, which is a UI demonstration, and which is a computed analysis? Do captions distinguish simulated hand movement from EEG-controlled movement? Do plots preserve axis units and condition labels?
reproduction checklist: record the relevant commit, source data identifier and command near each figure. Those details are proposed documentation improvements; the archived README changes do not establish that the complete provenance chain already exists.