The worksheet now uses SHERPA's published Figure 6.1 taxonomy: five error categories - Action, Checking, Retrieval, Communication and Selection - each driving its own set of credible error modes (A1 to A10, C1 to C6, R1 to R3, I1 to I3, S1 to S2). Selection, previously missing, is now present, and the mode list cascades from the chosen category rather than offering a flat non-standard list. Older assessments are migrated automatically as they load, so existing analyses map onto the correct canonical codes with no manual rework.

Each error now carries a dedicated error description (the form the error takes) and a written consequence, both previously blended into other fields. Remedial strategies are captured under the four standard headings - Equipment, Training, Procedures and Organisational - in place of a single free-text box. The Word, PDF and HTML reports render the new fields throughout, so an exported analysis reads as a complete SHERPA worksheet.

Criticality is now graded Low, Medium or High in parallel with probability, matching the method's ordinal treatment, and the consequence is captured as written analysis rather than a bare 1 to 5 number. Existing "critical" flags are carried across automatically on load, and the Extended (SHERPA+) mode keeps its numerical consequence for quantified scoring alongside the new criticality rating.

The separate Error and Error Mode columns are combined into a single Error Mode cell. A grouped picker dialog lets the analyst choose the category and its specific mode together from the full taxonomy, with each code shown in the cell and its full description on hover. A parkable, always-available reference window presents the complete Figure 6.1 taxonomy read-only, so it can sit beside the worksheet while a pre-filled assessment is reviewed.

Worksheet steps can be imported directly from a linked Hierarchical Task Analysis and kept in sync as the HTA grows, and a "View HTA" action opens the linked diagram with each analysed node badged by the worksheet row that covers it. A restyled grouping-tab strip, matching the ErgoSCTA Task Analysis editor, lets analysts carve a large study into named groups - by step range or hand-picked operations - with the groupings saved alongside the document.

Opening a project now shows an honest, on-theme busy indicator while it loads, and the heaviest part of the load has been moved off the interface thread so the indicator animates rather than freezing alongside everything else. Opening a larger project no longer looks like the application has stalled. The indicator appears on every open path, including auto-save recovery, and honours the Charlie, Harry and Classic themes.

The SHERPA worksheet now realises only the rows on screen, so opening a large session (60-plus steps, and well beyond) is near-instant instead of pausing for several seconds. Scrolling, editing, adding, reordering, grouping, reporting and Finalise & Sign all behave exactly as before, with the column headers now pinned above the scrolling rows.

Rolling backups and the periodic auto-save snapshot now live in a dedicated per-user folder instead of sitting beside your project, so a saved project leaves only the single project file in your working folder. All existing recovery behaviour is preserved - older files are still read from their previous location - and legacy projects are decluttered automatically the first time they are saved under the new scheme.

Building on the platform-wide isolation work in 1.3.1, the session layer is now on the same structural footing as the rest of the application. No tool's document, recording context, cached result or heatmap from one project can survive into another when projects are switched within a single session, and the mechanism is built so the clearing list can no longer quietly fall out of step as new tools are added.

Corrected a data-serialisation mismatch that could cause intervention-marker records to reload blank, distorting intervention analytics. Affected durable data paths now round-trip through a single canonical format, so records reload with their correct dates and values. The fix is precautionary as well as corrective: related persisted paths were aligned to the same canonical format to prevent similar drift in future.

Every change in this release - the canonical SHERPA taxonomy, the new worksheet fields, ordinal criticality and the relocated backups - is additive and backward compatible. Existing assessments and older project files continue to open cleanly, with legacy data brought up to date on load and no migration step required. The project recovery net is deliberately untouched: the atomic save mechanics are unchanged and older backups remain readable from their previous location.