Changelog

Performance & Reliability

Large Data Paste Handling

• Pasting huge Excel tables no longer freezes the application. Previously, importing journals with hundreds of thousands of rows would hang the browser for many minutes. The new optimized parser and incremental processing keep the workpaper responsive throughout the entire import.
• Massive improvements to memory usage for large data imports. The system now samples data intelligently rather than processing every cell upfront, preventing the browser from running out of memory on very large pastes.
• Smart confirmation dialog for very large pastes. Pastes exceeding 200,000 rows now show a recommendation to use the file upload feature instead, which handles that scale more efficiently without blocking the UI.

Data Integrity

• Fixed cell splitting when pasting Excel data with line breaks. Previously, cells containing quoted text with embedded newlines would incorrectly be split into multiple rows during import. This now works correctly.
• Single-column table pastes no longer split valid wrapped cells. Previously, wrapped cells in single-column tables would be broken into extra rows during import.

Authentication

• Users can now log in with the updated organization authentication domain. The organization's recent authentication system update is fully supported, so all accounts can access the platform without interruption.

Workpaper Sampling

• Saving selection data now generates a PDF sampling report alongside the Excel file. Click "Save Data" after setting up a Random, MUS, or Journal Entry sample and a formatted report is produced automatically. It includes a summary table, population breakdown chart, and method-specific parameters — plus the per-test hit-count table and Benford's Law histogram for Journal Entry testing (when that test is enabled).
• Sampling reports are bundled into all workpaper exports. Every export format — Workpaper Only, Files + Workpaper (.xlsx), and Files + Workpaper (.zip) — automatically includes the relevant sampling sheets and PDF reports.
• The Benford's Law histogram is now interactive. Click the chart to expand it to fullscreen for a closer view. You can also adjust how many digit pairs are flagged as anomalous directly from the fullscreen view.
• When you reopen the selection data dialog to edit an existing random sample, your previous filter settings are preserved.

Co-Audit Chat

• Edit existing workpaper columns without deleting and recreating them. You can now ask Co-Audit to update an existing column's name, prompt, data type, or pass/fail icons — the change applies in place, preserving the column's position and history.
• Choose how AI columns analyze files per row. When asking Co-Audit to add an AI prompt column or testing attribute column, you can specify whether the AI should evaluate each file individually or summarize across all files for a given row.

File Processing

• Very large spreadsheet and CSV files now process correctly. Files such as journal entry exports with 50,000 or more rows were previously rejected as too complex to read. These files now upload and extract text successfully.
• CSV files with accented letters, special characters, or files created with European regional settings now process correctly. Files that previously failed to upload due to encoding issues will now work.

Workpaper Automations

• Running all automations on a workpaper is more reliable. A performance issue that caused automation runs to fail silently on large workpapers has been resolved.

Task Management

• Save any task as a reusable template from the breadcrumb menu at the top of any task page (auditor only). Set a name, description, visibility (Personal, Firm-wide, or Global), and tags. The dialog previews the task's structure and warns you if a template with the same name already exists, giving you the option to overwrite.
• Client users can submit their completed workpaper for auditor review from the task overview page. If the task is already in review, they can return it to To Do.

Other Improvements

• Keyboard shortcut for task search. Press Ctrl+F (or Cmd+F on Mac) on the engagement board to open the search bar without clicking the icon.
• Edited comments are now marked. Comments that have been modified show an "Edited" label; hovering over it shows the exact date and time of the last edit.
• Drag files directly onto the engagement library panel to upload without opening the file picker dialog.
• ZIP files uploaded to the engagement library are automatically extracted — each file inside is added as a separate document.

Tick Mark Improvements

• Combined footing/crossfooting badges in exported PDFs: When a number carries both a footing tick mark and a crossfooting tick mark, the annotated PDF now displays them together as a single badge (e.g. "F-1·CF-2") rather than printing one on top of the other. Exception marks appear before validation marks within the badge, consistent with the order shown on the cover sheet.
• Comment icon placement fixed: Comment icons in annotated PDFs now correctly sit beside wider combined tick mark badges without overlapping them.

Manual Annotation Tick Mark Controls

Clicking a lasso-drawn annotation on the PDF now opens the same context menu used by auto-detected numbers. From this menu you can apply tick marks, toggle between Suggested and Confirmed, remove an association, or delete the annotation entirely — the same controls, in the same place, for both detected and hand-drawn annotations.

Math Footing Calculator

Sign (+/−) changes now stick. Toggling a number's polarity in the footing panel would revert to the original sign on the next load. The toggle is now saved correctly.

Pressing Escape cancels a Stated Result edit. Previously, pressing Escape while typing in the Stated Result field committed the in-progress value instead of discarding it. Escape now restores the original value, as expected.

Stated Result is editable in automatic footing mode. Auditors can now correct an OCR-read Stated Result — such as a misread sign — when working in automatic mode, not only in manual mode.

European-formatted numbers calculate correctly in manual footing. Numbers using dots as thousand separators (for example, TEUR format 153.063) were misread as decimal values, producing incorrect sums. Manual footing now interprets these numbers the same way automatic mode does.

Operand order follows document reading order. Numbers in the footing panel now appear left-to-right within each row. Column crossfootings sort right-to-left across the page, matching the direction auditors naturally work.

Unreadable OCR values are clearly flagged. When document processing cannot interpret a number, the footing panel now shows an amber "Enter value" prompt instead of silently defaulting to zero — making it obvious where manual input is required.

Validation errors in the footing panel are now visible. Error messages (such as "Please enter a valid number") now appear as on-screen alerts. They were previously silent.

Reliability

Large multi-sheet Excel files now process successfully. Files with many sheets and a high volume of cells were previously causing the processing job to exhaust available memory, failing the entire upload. These files now complete processing without issue.

Bug Fixes

• Clicking a citation that references a different file now works correctly. If you were viewing one document and clicked a citation pill pointing to a different source file, the cited document failed to open. This is now fixed — clicking any citation pill switches to and highlights the correct source document, regardless of which file you were previously viewing. This applies to Word and Excel documents in addition to PDFs.

• Citation pills in certain workpaper cells now resolve correctly. Some citations in workpaper cells appeared as broken, unresolved references rather than clickable pills. These now correctly display as clickable pills and navigate to the cited page in the source document when clicked.

• Agentive-generated Excel workbooks are now consistently recognized by the Excel add-in. Workbooks produced by Agentive's AI automation are now always compatible with the Agentive Excel add-in when opened on your desktop, even if no citations were added.

Document Viewer

• Consistent empty state when no file is associated: Clicking a cell that has no linked document now shows a clear message — "No file selected" (with a prompt to pick one from the panel below) or "No files uploaded to this datasource yet" (with an upload shortcut) — instead of auto-loading an unrelated file from the datasource.
• Visual marker for rows without a matched document: In datasources where some rows didn't receive a matched document, the extraction cells for those rows now show a blue dash on a gray background, making it immediately clear that AI extraction won't run on that row.
• Viewer panel width no longer jumps: The document viewer holds its width as you move between different cell types — no more resizing when switching from a match cell to a retrieval cell and back.

Cell Detail Panel

• Section state remembered as you navigate: Collapsing a section (Answer, Reasoning, Activity, and others) now keeps it collapsed as you move between cells in the same session. Reloading the page resets everything back to fully expanded.
• Tick mark cells always reachable: Cells that have a tick mark applied but no associated document now correctly show the Cell tab, so you can view or update the applied tick mark without it disappearing.

Shared Document Groups

• Per-task matching mode: When multiple workpapers share a document group, each workpaper can now independently choose AI matching or Custom Rules matching — changing the setting in one workpaper no longer affects the others.
• Automatic ownership transfer: If the workpaper that originally created a shared document group stops using it, ownership automatically passes to another workpaper still using the group so it stays accessible with no disruption.
• Switching groups preserves settings and workflow: When you replace an attached shared document group with a different one, the current workpaper's matching configuration carries over and any connected downstream datasources are automatically re-linked — the workflow diagram stays intact.
• Full access for importing workpapers: Workpapers that share an existing document group can now comment on cells, change matching mode, and trigger runs — previously some of these actions silently applied to the original owner workpaper instead.
• Workflow diagram arrows always visible: Parent-to-child arrows now appear correctly for all datasources regardless of which matching mode each one is configured with.

AI & Testing Attributes

• Collective testing attribute results corrected: Testing attribute columns set to Collective mode now correctly produce one shared pass/fail verdict for the entire record group rather than one result per record.
• Testing attributes on document-only datasources: Testing attribute columns now run correctly on datasources that contain only uploaded files (with no parent table), producing one result row per uploaded file.

New

• Automatic fiscal year detection: When you run Prior Year Consistency, the pipeline now reads the cover pages of your uploaded CY and PY documents to identify the correct fiscal year automatically. This means PY Consistency runs against the right year even when the engagement's audit period hasn't been configured. If the documents don't contain a clear dating phrase, the pipeline falls back to the engagement period as before — and replaces the auto-detected period whenever you swap out the uploaded files.

• Editable "Current year end" date in Prior Year sidebar: The fiscal year-end date used for PY Consistency is now visible and editable directly at the top of the Prior Year sidebar. You can correct the auto-detected date when needed. Clearing it re-enables automatic detection on the next run, and an info tooltip explains what the date controls.

• Combined IC + PY tick marks in exported PDFs: When a number carries both an Internal Consistency and a Prior Year tick mark, the exported PDF now shows them as a single readable label (e.g., "PY-7·IC-3") rather than two overlapping marks that hide each other.

Improved

• IC groupings are more precise: Numbers are now grouped only when they represent the same financial fact — same concept, same period, and same scope. Figures that appear in adjacent comparative columns (the same line item in different years) are correctly kept as separate items rather than surfaced as potential tie-outs.

• Empty subsections show "None" instead of blank headers: In the IC, Prior Year, and F/CF sidebars, subsections with no items now display a muted "None" label. It's immediately clear that the section has been reviewed and found clean, rather than leaving an empty collapsible header that appears to have content.

• Shared document group confirmation is now inline: Replacing a shared document group with another now shows a single inline summary of what will change — including which task will handle client uploads going forward — without opening a separate modal.

• Export error files include clear next steps: When a multi-file PDF export fails for one or more files, the error file inside the downloaded ZIP now opens with plain instructions for what to do, including a ready-to-copy message you can send directly to support.

Fixed

• "Check all" and "Flag all" buttons in IC and PY sidebars now work: Clicking Check on the Consistent bucket or Flag on the Flagged bucket was silently doing nothing. Both buttons now correctly approve and confirm the relevant suggested items.

• Prior Year loading indicator no longer spins indefinitely: After running or re-running Prior Year Consistency, the loading spinner in the sidebar clears as expected once the run finishes.

• PDF export correctly reports partial failures: When exporting multiple annotated PDFs and some files fail, the notification now reads "Export completed with errors" (or "Export failed" if all files failed) instead of incorrectly showing a success message.

• "Run everything" correctly identifies columns that have never been run: The run modal now properly recognizes new custom columns that haven't been executed yet, so the option to fill empty cells is available as expected.

• Activity summary emails deliver reliably: Fixed a bug where certain removed tasks could cause the daily activity summary email to fail to send.

• Shared document group data stays isolated to your task: In workpapers using shared document groups, certain data panels now correctly show only the current task's entries rather than mixing in entries from other tasks that share the same group.

Shared Document Groups

• Smarter import flow when client requests are involved: Switching a document group to a different shared source now handles every combination of client-upload ownership with the right confirmation experience. When both the current and target groups are owned as client requests by different tasks, a dedicated dialog appears — showing which task currently handles client uploads, what files will be affected, and what happens to client upload routing — before you confirm.
• See exactly which files will be permanently deleted before switching: When switching to a different shared document group would permanently remove files, the panel now shows the exact file count upfront. The confirm button stays disabled while this check runs, so you can never accidentally confirm before the full impact is visible.
• "Import document group" toggle: The toggle in the document group editor panel is now labeled "Import document group" when the group is standalone, and "Shared document group" once sharing is active, making its current state clearer at a glance.

EU and German Document Accuracy

• Fixed: EU financial values ending in zero cents were being dropped: In EU-format tables — such as German balance sheets — values whose cents ended in zero (for example, 15.375,50 or 16.991,00) were being silently excluded during extraction. This affected roughly 30% of numbers in affected tables, causing footings to fail and numbers to appear missing. These values now extract correctly.
• Fixed: prior-year comparisons could mis-route for round amounts: A related issue caused the year-over-year comparison logic to silently treat all values as equal for certain round figures, which could misdirect reconciliations across periods. This is now resolved.
• German section references and legal citations no longer treated as financial numbers: German document markers — such as "Anlage 3", "Kapitel 5", "§ 322 HGB", and numbered headings — are now correctly excluded from financial number extraction, reducing noise in AI analysis on German audit reports.
• German fixed asset schedules no longer incorrectly combined: The system was sometimes merging separate sections of a fixed asset schedule (Anlagespiegel) that document different financial events — for example, acquisition costs on one page and accumulated depreciation on another — into a single table, dropping data in the process. These sections now remain separate.
• Fewer false inconsistency flags on comparative-period columns: The inconsistency checker was occasionally raising exception marks on comparative-period columns based solely on shared row position, even when the values are expected to differ across periods. This has been corrected.
• Improved extraction for EU headerless and multi-block schedules: Table extraction quality improved for EU-format financial statements that span multiple sections without standard column headers.

Processing Reliability

• PDFs with complex graphics now process reliably: PDFs containing many vector-graphic elements — such as diagrams or heavily formatted pages — were causing OCR processing to fail entirely for that file. These files now process normally, with all text and numbers extracted as expected.

Bug Fixes

• Deleting tasks with many attached files now completes reliably. Previously, deleting a task with a large number of files could fail partway through, leaving the task in an inconsistent state. The deletion now runs to completion every time.
• An unexpected error message on certain page transitions is resolved. Some users were occasionally shown an "unexpected response" error when the app redirected them between pages — for example, after a session change or when accessing a page they don't have permission to view. These redirects now work smoothly without showing an error.

Bug Fixes

• Sum calculator restored in the cell detail panel: When working with cells that sum values across a document, the yellow Σ total display and calculator had disappeared from the right-hand panel. It's now back — clicking into a sum cell shows your running total, and you can adjust the sign of individual components or remove them directly from the panel.

• Match automations running again: On some workpapers, the automated matching process was silently stopping before completing. This is now resolved — match columns run to completion as expected.

• AI chat handles interrupted sessions more reliably: Two separate issues could cause the AI chat to stop responding — one when a tool action was cancelled mid-run, and one when a session was interrupted before completing. Both are fixed; resuming a chat after an interruption now works correctly.

Removed

• "Create by Procedure" is no longer available: The task creation flow that accepted a procedures document and generated a suggested workflow from it has been removed. Tasks can still be created from templates, from scratch, or via the Financial Statement Review flow.

Improvements

• Applying a template is now instant. The intermediate document group mapping screen has been removed from the template apply flow. Selecting a template and clicking Apply creates the task immediately — no extra confirmation step required.

• Email invites accept any paste format. When adding team members or clients to an engagement, you can now paste addresses in any common format: space-separated, comma-separated, one per line, or directly from Outlook ("First Last <email@firm.com>"). Invalid addresses are flagged inline before the invite is sent, so you can fix them on the spot instead of hitting a generic error after the fact.

• Foot/Crossfoot calculator lists operands in document order. Values in the F/CF calculator are now shown in the order they appear in the PDF — page by page, top to bottom — making it straightforward to trace each number back to its location in the source document.

Bug Fixes

• "Apply Valid" / "Apply Exception" no longer fails after a tick mark is removed. Re-applying a footing result to a number that previously had its tick mark removed now succeeds as expected, instead of showing a "Failed to finalize calculation" error.

• AI chat panel no longer goes blank during streaming. An issue that caused the chat panel to crash and stop rendering while a response was being generated has been fixed.

• Workpaper page is stable on long-running engagements. An issue causing the workpaper page to crash on engagements with a long activity history has been resolved.

Bug Fixes

• Automations now run correctly on imported document groups. When a document group is shared with your workpaper from another workpaper in the engagement, running automations — including testing attributes and run-all — now operates on your workpaper's data. Previously, these operations could silently target the original owning workpaper and produce no output.

• Workpaper grid no longer shows blank rows after automation runs. If your workpaper uses an imported document group, running a full automation could cause rows to disappear from the grid. This is now fixed.

• Match accept operations apply to the correct workpaper. Accepting file or record matches on a shared document group now correctly records results under the workpaper you are working in, not the workpaper that owns the document group.

• Live updates now refresh the correct workpaper. When an automation clears or updates a shared document group, the refresh now appears on the workpaper you are viewing rather than the owning workpaper.

AI Improvements

• German, Italian, and Spanish financial statements: Internal Consistency analysis now correctly identifies and ignores non-financial numbers in EU-language documents — page numbers, footnote references, phone numbers, dates, table-of-contents markers, and identifier abbreviations are filtered out in German, Italian, and Spanish, matching the coverage already in place for English. Auditors working on EU-language engagements will see cleaner, more relevant IC candidates with less noise to review.

Bug Fixes

• Shared document groups now run correctly: When a task uses a document group shared from another task (the "Use existing" option), workpaper automations were sometimes pulling in evidence from the wrong task or returning no results at all — leaving cells blank after a successful run. Both issues are resolved; shared document groups now process accurately.

• Reference files visible after sandbox provisioning: Reference files attached to template engagements were not appearing in the provisioned sandbox copy. This is now fixed, and engagements already affected have been corrected automatically.

• FSR re-run progress displays reliably: When re-running a single Financial Statement Review step (internal consistency, prior year, or foot/crossfoot), dismissing the progress notification before it finished could cause subsequent reruns to show no visible progress. Fixed — the progress card now appears correctly on every rerun.

EU Number Formatting

• European-formatted numbers are now fully supported across math validation and financial statement review. Firms using commas as decimal separators and periods as thousands separators (e.g. "1.234,56") will see numbers parse, foot, and display correctly throughout workpapers and the FSR review sidebar — previously these were misread or shown in US format.
• Multi-group European integers like "1.234.567" are now correctly recognized as whole numbers rather than being truncated.

Task Creation

• After creating a task from a template, you're taken directly to the most relevant page — the Workpaper for templates that include data sources, the Review page for document-review tasks, and the Overview for everything else. Previously, template-based tasks would sometimes drop you on the wrong page.
• A one-time "New: Reuse groups in templates" tip appears the first time you reach the template reconciliation screen, with a matching tip the first time you use "Use Existing" in the workflow sidebar. Both dismiss with a single "Got it" click.

Workpaper

• The Problems panel no longer shows a misleading entry before analysis has been run. Every new workpaper used to show a permanent informational row prompting you to run analysis — even when nothing was actually wrong. That entry is removed; the panel now only surfaces real issues.
• The Problems panel is visually cleaner: Fix with Co-Audit buttons on category rows now appear on hover (matching the per-row behavior), and the severity count badges next to category names have been removed.
• Automations run on shared document groups now show their progress and results correctly in the chat. Previously, the progress card could appear empty and results might never arrive in the chat when a data source was shared across tasks.

PDF Viewer

• The annotation drawing tool remembers your last-used color between sessions. The default color is now grey (previously red), and your chosen color is restored the next time you open a file.
• The FSR sidebar now shows "No findings yet" when no results are present, more accurately reflecting that running a test will produce output.

New Features

• Sticky artifacts bar in Co-Audit Chat: A persistent bar above the chat input now surfaces every downloadable file the AI has produced in the current conversation — Excel workbooks, Word documents, PDFs, ZIP archives, and other exports. The bar collapses to a count badge when not in use and expands to show labeled, colored pills for each file with one-click download. It stays visible as the conversation grows, so you can reach any generated file at any point without scrolling back through earlier messages.

• Draw proof marks directly on PDFs: A new drawing tool in the document review toolbar lets auditors mark up PDF pages with lines, arrows, and accounting double-underlines. Choose a stroke color from a 9-color palette — the toolbar icon tints to show your active selection. Click any mark to select it; a delete button appears for your own marks. Marks are stored per file, visible to all viewers, and included in the annotated PDF export.

Improvements

• Problems tab refreshes automatically: The Problems tab in the workpaper now updates on its own as soon as AI analysis finishes running. You no longer need to reload the page or re-run the check to see the latest results.

• Post-run fix suggestions back in action: After running automations on a workpaper, the Co-Audit assistant once again provides grouped, prioritized suggestions for the issues found — a behavior that had been inadvertently dropped in a recent update and is now restored.

• Document group removal is clearer and safer: Removing a document group from a workpaper now clearly separates "remove from this workpaper only" from "delete entirely." When you are the last workpaper using a group, the confirmation dialog shows exactly how many headers and values will be affected and includes an explicit option to delete the group and all its files — so nothing is permanently removed unless you choose it.

• AI-generated PDFs now download directly: PDF files produced by the AI assistant download to your computer immediately, consistent with how Excel, Word, and ZIP artifacts behave. Previously they opened in the browser instead of downloading.

• More accurate financial table extraction: Balance sheets and similar documents where a row label appears in multiple adjacent columns are now extracted correctly. Numbers are matched to the right column positions rather than being shifted one column off.

• Workpaper template reuse picker scrolls correctly: When applying a template on an engagement with many document groups, the reuse picker now scrolls cleanly through the full list. Sample datasource rows in the picker are correctly locked to "Create a new group" (with a tooltip explanation) since that type of data cannot be reused across tasks.

Bug Fixes

• Document viewer no longer carries over between pages: A document opened in the workpaper viewer no longer persists in the side panel when navigating to the Overview, Memo, or Chat page. Each page now opens with a clean state.

• Comment threads fully visible on active tasks: On tasks with high volumes of file activity, comment threads were appearing empty in the sidebar even though replies existed. All threads now display in full.

• AI file creation more consistently available: Fixed a recurring issue where the AI's ability to produce files (Excel, Word, PDF) could become unavailable for an extended period following spikes in cloud infrastructure pressure. The system now cleans up proactively to prevent capacity from being exhausted.

Fixes

• Internal Consistency no longer flags figures from different time periods as potential tie-outs. When the same line item appears in side-by-side comparative columns — for example, current-year and prior-year balances on the same row — the tool now recognizes these as two separate facts that are expected to differ, and leaves them as individual items rather than surfacing them as matches to investigate.

• Zeros, dashes, and blank cells are no longer clustered across unrelated schedules. These values typically mean "no activity" or "not applicable" for each individual cell. The IC tool now treats them independently rather than grouping them as apparent matches across equity rollforwards, fair value schedules, and other multi-column layouts — eliminating large clusters of spurious findings auditors would otherwise need to dismiss.

• Numbers throughout multi-section tables are now attributed to the correct column. In tables with distinct upper and lower halves — such as a balance sheet divided between an assets section and a liabilities section — figures in the lower section were sometimes tagged to the wrong column heading. This is now resolved, improving the accuracy of foot and crossfoot validation results for these tables.

New

• Empty workpaper onboarding: when a workpaper has no columns yet, the grid now shows a clear "Set up columns" message with two options: a Draft with Co-Audit button that opens the Co-Audit chat and immediately asks the AI to propose columns from your task context, or an "Or configure manually" link that opens the workflow sidebar. The previous placeholder — fake column names like "Invoice #" and "Vendor" — has been removed.

Workpaper

• Cell panel follows your focus: clicking a different cell while viewing the Problems or Document tab now automatically switches to show that cell's content. Previously you had to click back to the Cell tab manually.
• Consistent review action labels: the Validate, Flag, and Lock pills now use matching past-tense labels ("Validated", "Flagged", "Locked") across all column types. These actions are hidden entirely on columns containing pasted sample data, where AI judgments don't apply.
• Problems tab actions in the tab strip: the Run All and Fix-with-Co-Audit shortcuts now appear on the right side of the tab strip when the Problems tab is active, rather than inside the panel body.
• Deleting files is more reliable: deleting a file that had math tick marks attached no longer fails partway through — the cleanup now completes successfully.

Financial Statement Review

• Spell Check and Grammar Check share findings: the two checkers now coordinate so Grammar Check defers to Spell Check on overlapping issues (possessives, punctuation, terminology). The same underlying problem no longer appears flagged twice.
• Correct wrong footing totals directly: when OCR reads an incorrect value in a row total or column cross-footing, you can now edit it in place — the same inline correction experience that was previously available only for manually-drawn footings.
• Findings visible immediately in the F/CF sidebar: findings are no longer grouped behind collapsed page headers. All flagged and consistent items are visible as soon as the sidebar opens.
• Math calculator errors stay visible: when saving a footing calculation fails, a persistent error banner with Retry and Revert buttons appears inside the panel, keeping your edits intact. Previously, only a brief notification appeared and the panel appeared saved.
• Fewer false-positive Internal Consistency suggestions: the IC checker no longer flags schedule entries labeled "No." (as in "Special District No. 2") as financial figures, no longer groups debt maturity schedule rows as a consistency tie-out, and now requires specific evidence for each individual figure rather than a single general rationale for an entire group.
• SC/GC PDF overlay shows only the selected finding: previously every finding in a Spell Check or Grammar Check session was highlighted on the PDF simultaneously. Now only the finding you click is highlighted.
• Math sidebar scrolls less aggressively: navigating between occurrences in the math sidebar no longer jumps the entire section to the top — the view scrolls only enough to keep the active row visible.

PDF Viewer

• Ctrl+scroll zoom no longer snaps to page 1: using Ctrl+mouse wheel to zoom out previously could drive the zoom to 0%, collapsing all pages and snapping the viewer to the beginning of the document. Zoom is now clamped to a safe range and page position is preserved.

Reliability

• Automation failures show clearly: failed automations — including Review Checklist, Spell Check, Grammar Check, and math analysis — now display as errors in the progress panel. Previously they could appear as successful or spin indefinitely.
• AI correctly reports when a file is temporarily unavailable: if a document becomes unreachable mid-search, the AI now tells you to retry rather than reporting "no matches found." If partial results were found before the interruption, those are returned alongside a clear warning.
• FSR pipeline runs more reliably on multi-section documents: a timing issue that could cause the Internal Consistency step to proceed without embeddings from all document sections — adding a several-minute delay with no visible explanation — has been resolved.

New Features

• AI-produced files in task chat. The Co-Audit agent can now build custom Excel spreadsheets, Word documents, and PDFs directly in the chat — not just describe what to do, but hand you the finished file. Ask for things like "build a loan schedule with a separate sheet per vendor" or "fill this form from our workpaper data" and a colored download chip appears at the end of the response. The agent can also read evidence and reference files already attached to the task when building its output.

• Move and Copy matched evidence. Right-clicking a matched row in the workpaper now offers a Move or Copy option under "Move to":

  • Move — removes the original match and creates it under the chosen parent row, with a preview of all downstream records that will be affected before you confirm.
  • Copy — adds a second association without touching the original, useful when the same evidence belongs under multiple parents.

• Run on empty match cells. Right-clicking an empty match column cell on a parent row now shows a Run option alongside "Pick document," so you can ask the agent to fill the match automatically rather than picking a file manually.

Improvements

• Template reconciliation is faster and clearer. When applying a template whose document groups share names with groups already in your engagement, the reconciliation screen now:

  • Lets you type a custom name when creating a new group (formula references update automatically).
  • Shows a "Client request · ID" chip next to any reuse option that's already shared with clients, and warns you that client uploads will continue going to the original task — so you know the upload ownership consequence before confirming.
  • Skips the screen entirely when there are no naming conflicts, jumping straight to a one-click Apply summary.
  • Highlights inline when two rows would resolve to the same name.

• First-run modal is simpler. When you open the Run dialog on a scope that has never been run, you now see just the scope description and a single Run button. The breakdown table and duplicate Re-run button only appear after something in the scope has been run.

• Remove Match is consistent everywhere. The consequence preview — showing how many downstream matched rows will be removed — now appears regardless of whether you use the grid right-click menu, the cell-detail panel, or the document-detail panel.

• Document group picker shows more context. The "Use Existing Document Group" panel now displays a file count badge on each group row, and task labels include the short task ID for faster scanning.

• AI artifact sessions start faster. The Co-Audit agent's file-building sessions now reach a ready state noticeably faster than before.

• Better AI model for EU users. The default AI model for auditors in the EU region has been upgraded to match the model used in the US.

Bug Fixes

• Co-Audit sidebar no longer splits in two. Opening "Fix with Co-Audit" from the Problems tab previously rendered a duplicate strip of recent chats below the chat panel. The chat panel now takes the full sidebar height as expected.

• Download chip appears once on saved chats. Reopening a chat that contained an AI-produced file used to show the download chip twice. It now appears exactly once.

• Template column setting for "run on no match" is now preserved. A per-column setting that controls whether the agent runs when no evidence match is found was being silently dropped when saving or applying templates. It now round-trips correctly.

New

• Document tab in the workpaper file viewer: When a file is open in the viewer, a new Document tab shows all the file's details in one place — its classification (editable inline), which match group it belongs to (Unmatched, Match Unlocked, Match Locked, or Excluded, all changeable from a dropdown), related page ranges from the same document, the AI's notes and reasoning, and a per-file activity history.

• Match section inline in the cell detail panel: When you click a match cell, the file picker now appears directly in the cell detail panel below the grid, split into "Matched" (files already linked to this row) and "Other" (remaining files in the datasource). No need to switch between panels to manage a match.

• Direction toggle in the match panel: A toggle in the match panel header lets you flip between two perspectives: "files view" shows which documents you can match to the focused row; "rows view" shows all rows in the datasource grouped by whether they're already matched to the file you're currently viewing in the viewer.

• Move and Copy for matched records: Right-clicking a match now offers explicit Move and Copy choices. Move re-parents the match to a new row and removes it from the original — a confirmation dialog shows you the full count of records that will be affected before you commit. Copy adds a second match at the new location while leaving the original intact.

• Run AI on empty match cells: Right-clicking an empty match cell now shows a Run option alongside the manual file picker, letting you ask the AI to attempt a match automatically.

• Run confirmation dialog: Every Run button now opens a confirmation dialog showing a per-document-group breakdown of what will be run — matches, custom column cells, and testing attributes. You choose between Run (fill only empty results) and Re-run (overwrite all non-locked, non-reviewed results). On the very first run of a fresh workpaper, when nothing has been filled yet, the dialog collapses to a single Run button with no breakdown.

Improvements

• Tick marks can be re-applied after removal: Removing a tick mark and then re-applying the same one to the same number no longer causes a conflict. The removed annotation is preserved in the history rather than being permanently deleted.

• Client setup error message: If a new client's practice resources don't finish setting up within five minutes, the loading spinner now transitions to an error message directing you to contact support instead of spinning indefinitely.

Bug Fixes

• Workpaper grid cleanup after re-run: After running all automations, empty placeholder rows no longer remain visible in the grid until the page is reloaded — they clear automatically.

• "Fix with Co-Audit" no longer splits the chat panel: Opening "Fix with Co-Audit" from the workpaper no longer causes recent chat history to appear alongside the active conversation, splitting the panel in two.

• Sandbox engagements don't generate email notifications: Activity summary emails and @mention notifications are no longer sent for sandbox (practice) engagements.

• Partial match Move is clearly reported: If a match Move creates the new association successfully but cannot remove the original, you now see a specific message explaining what happened and how to finish the move manually.

Improvements

• Add Value stays accessible during operand assignment: When building a foot/crossfoot calculation by lassoing numbers one at a time, the Add Value button now remains visible even while a previous number is waiting for its row and column to be identified. You can keep adding numbers to the calculation without the button disappearing mid-flow.

• Row footing and column crossfooting are tracked as separate annotations: Tick marks for row footing and column crossfooting on the same number now coexist independently. Placing a column crossfooting no longer silently upgrades an existing row footing to Confirmed — both marks are preserved and reflect their correct direction.

Bug Fixes

• Task template description is now optional: The Save button in the Save-as-Template dialog is no longer disabled when the description field is empty. Templates can be saved without a description.

• Template save errors are now specific: When saving a template fails, the dialog now displays the actual reason (for example, if the source task is no longer available) instead of a generic "Please try again" message.

• Deleted-chat links no longer cause errors: Opening a link to a chat that was deleted, or restoring a session where the last active chat no longer exists, no longer produces an unexpected error.

New Features

• Workflow diagram in workpaper: A new interactive diagram in the workpaper sidebar visualizes all datasources as cards with connecting edges. Each card shows its type (Table Data, Client, or Auditor), column and file counts, and yellow warning banners when selection data or files are missing. Action buttons at the bottom let you run all automations or drill into per-datasource options — including a "Fill Empty" mode that skips cells that already have values.

• Context picker in task chat: An attachment popover now lets you choose exactly which files to include in your chat message. Files are grouped by datasource with "Select all / Deselect all" per group, a filename search, and a count badge. In task chat, a "Paste Selection Data" action opens the Add/Edit Selections modal directly from the chat panel — no need to navigate to the workpaper first.

• Chat can navigate directly to a sampling form: The Co-Audit Agent can now deep-link you to the workpaper and automatically open the selection data modal pre-set to a specific sampling method (Random, MUS, High Value, or JE Testing) — so following an AI recommendation takes you straight to configuration.

• Workpaper onboarding now offers two starting paths: When setting up a new workpaper, you can choose "Add Selections" (opens the import modal right away) or "No Selections" (proceed directly to naming a document group), matching how you actually want to begin.

Improvements

• Clearer selection data controls: The workpaper datasource toolbar now shows an "Add Selection Data" button when no data exists, and an "Edit Selection Data" split button when data is present. The split button separates "Replace Selection Data" (clear and re-import, including changing headers) from "Append Rows" (add rows while preserving existing column structure).

• Done tasks visible by default on the board: Completed tasks now appear on the board by default. Use the "Completed issues" filter to hide them — opting in to hiding rather than opting in to seeing.

• Status change confirmation when selection data is missing: When moving a client-visible task to a status clients can see, and the task has no pasted selection data, a confirmation dialog now appears. It lists each affected task and requires an acknowledgement before proceeding — preventing accidental client exposure of requests without data.

• Library file preview improvements: The library now shows a graceful fallback for corrupt or unreadable PDFs instead of a blank error. Excel previews include sheet tab navigation so you can browse between tabs.

• Welcome email errors now show the specific reason: If sending a welcome email to an auditor or client fails, the notification now shows the specific cause (e.g., "on suppression list — contact support to remove") instead of a generic error.

New Features

• AI exception clustering on TICKMARK automations: when a TICKMARK attribute column automation completes, AI now automatically groups the exception rows (rows that failed the tickmark evaluation) into labeled exception groups (A, B, C, …). Re-running the automation preserves any cells you have already accepted or rejected, so your review work is never lost.

Bug Fixes

• PDF export: F and CF tickmarks now render correctly when opening exported workpapers in Adobe Acrobat.
• Paste Sample / Add Table from Excel: the import now validates the column mapping upfront and surfaces a clear error instead of failing mid-operation.
• Automation reliability: fixed an edge case where re-running an automation on empty cells would not register correctly, causing the run to silently stall.
• Tickmarks: resolved a race condition that could produce errors when two tickmarks were created on the same workpaper in quick succession.

Improvements

• New Engagement form: The Create button now stays disabled until the engagement name is entered, making it clear when the form is ready to submit.
• Activity Feed filters: Filter labels have been updated to plain language — People, Activity Type, and Client replace the previous technical labels, making it easier to find what you're looking for.

Bug Fixes & Improvements

Automation

• Unmatched file records are now excluded from Run All across every column: Previously, the exclusion of unassigned files (records the match agent couldn't link to any upstream selection) was not consistently applied to all column types in match-result datasources. It is now enforced universally — every column in a Run All pass skips these records, preventing degenerate or meaningless results from propagating into your workpaper. Root datasources and lineage-scoped runs are unaffected.

Reliability

• Improved resilience during high-load automation runs: Internal database connection handling now retries more aggressively across a wider window, reducing the chance of transient connection errors during large or concurrent workpaper automation runs.

New Features

• PDF citation highlighting in chat: When the AI references a value from a source document, it can now pin the exact region directly in the document viewer — not just quote the text. Click the citation to jump to the highlighted area in the PDF. Available in task chat once document context exists (file reads, attachments, or extraction results); available in Co-Audit Chat whenever attachments are present in the conversation.

Improvements

• Engagement variable context management in chat: In long AI chat sessions, previously fetched engagement variable values are automatically compacted out of the AI's working context and replaced with a lightweight reference — keeps sessions running cleanly without hitting context limits. The AI re-fetches the value automatically if it's needed again.
• New Client button on client detail page: You can now create a new client directly from a client's detail page header without navigating back to the client list.

Bug Fixes

• Match column "matched manually" label corrected: The "This file was matched manually" tooltip on workpaper match cells now only appears when a file was genuinely assigned by hand. Files matched by the AI agent or rules-based matching no longer carry this label.
• Duplicate client name validation: Attempting to create a client organization with a name that already exists in your firm now surfaces an inline error ("A client with this name already exists") immediately, rather than failing without clear feedback.

AI Improvements

• Co-Audit Agent skips approval for empty datasource deletion: deleting a datasource with no files, columns, or records no longer surfaces a confirmation card — the agent proceeds immediately. Approval is still required when the datasource has content.
• Workpaper design anchors on purpose and source: when building or modifying workpaper structure, the Co-Audit Agent now grounds every structural decision in the audit objective and the documents the auditor will examine, inferring both from context (prompt, task title, memo, existing datasources). If context is insufficient, the agent names what's missing before proceeding — no more guesswork on downstream choices like datasources, columns, or match instructions.
• Automatic cleanup of the "Requested Documents" placeholder: the default Requested Documents datasource created on new tasks is now treated as a placeholder. If it's empty when the agent is designing the workpaper, it's renamed to fit the design or deleted — never preserved as-is just because it exists.

File Management

• Bulk Move to datasource: select multiple files in the file explorer and move them all to a different datasource in one action via the Actions dropdown or right-click menu. If any selected files have protected records, a confirmation dialog shows the count before proceeding.
• Upload cell address tooltip: hovering any upload cell now shows a cursor-following chip indicating exactly where dropped files will land — per-row cells show "Selection #N → DatasourceName", the bulk upload cell shows "Bulk upload → DatasourceName". The tooltip automatically hides during active drags so it doesn't compete with the table-level drop indicator.

File Processing

• Accurate citations in large PDFs: large PDFs processed via the Lambda pipeline now correctly capture per-page text and graphics across all pages, fixing citation bounding boxes that were off or missing on later pages.
• Clear permission error on upload: when a file upload fails because the user doesn't have access to the workpaper, the file tooltip now shows a distinct permission-denied error instead of a generic failure message — making it immediately clear the issue is authorization, not a processing problem.

Bug Fixes

• Excel export — matched filenames: MATCH and MATCH_AGENT datasource cells now show the matched filename in the export instead of appearing blank.
• Excel export — redundant column removed: the row-level "File (Source)" column has been removed from Excel exports.
• Task creation modal — cursor position: typing in the description field of the task creation modal no longer causes the cursor to jump to the end of the text.

UI

• Request table redesign: upload cells, column headers, footers, and drag-and-drop interactions have been refreshed with updated styling and cleaner UX.
• File explorer folder icons: datasource folder icons now open visually on hover and when the folder is expanded, making the tree state easier to read at a glance.

Improvements

• Roll forward no longer carries stale prior-year variable values: Engagement variables linked to workpaper columns (synced variables) are now left blank when rolling an engagement forward — they populate automatically when the destination workpaper runs. User-set variable values continue to carry forward as before.
• Re-running roll-forward is safer: Re-running against an existing destination no longer overwrites synced variable values that the destination workpaper has already populated.
• Roll Forward dialog always pre-populates from completed engagements: Navigating to Roll Forward from a prior-year engagement marked as Done now correctly pre-selects the client and engagement, even if engagement data is still loading.

AI Chat

• Look up engagement variables in Co-Audit Chat: The AI can now list all engagement variable names and retrieve the current value of any specific variable (e.g. "What's the materiality threshold?"). Works in both task chat and standalone chat.

Bug Fixes

• 404 pages redirect to your engagement: Navigating to an unknown or mistyped URL no longer leaves you on a dead-end error page — authenticated users are automatically redirected to their most recent engagement.

Bug Fixes

• Financial Statement Review page is now properly restricted to auditors — clients who navigate to a review URL are redirected automatically rather than encountering an incomplete or unauthorized view.

New Features

• FSR PDF exports now open with a structured cover sheet. The prepended cover sheet provides a full tick-mark index in two sections — §01 lists every flagged exception with its mode, tick-mark identifier, description, and reviewer notes; §02 is a complete index organized by mode (PY → IC → F/CF). Every row is a clickable internal link that jumps directly to the matching page in the financial statements (and to the exact on-page location when a bounding box is available). The cover sheet also includes a masthead (client name, engagement title, reporting period) and a six-entry tick-mark legend, making the exported PDF self-contained for offline review.

Bug Fixes

• Cloned engagements: reference and library files are now accessible to all recipients. Previously, auditors provisioned via a sandbox clone were denied download and viewer access to reference and library files because file ownership still pointed to the source engagement. Files are now correctly scoped to the destination task, and access is restored immediately for all clone recipients — including non-admin auditors.

• Engagement board: disabled board pills now show real counts and smarter guidance. When the Requests or Tasks board is turned off in engagement settings, the pill now shows the actual task or request count instead of a hardcoded zero. If hidden tasks exist — for example, after cloning from an ALL-visibility engagement — the tooltip explicitly flags it: "This engagement has client-visible requests but the board is hidden. Enable it in engagement settings." When the count is zero, the original shorter tooltip is shown. Auditors can now tell at a glance whether there are tasks waiting on them vs. whether the board is simply unused.

• Automation progress cards are more accurate after network interruptions. On reconnect following a pod rotation, stale failure markers in the automation progress card are immediately replaced by the correct live status if the background retry has already succeeded — no more misleading red indicators on automations that actually completed. In the AI chat run card: a "Reconnecting…" pill appears during brief connectivity gaps so in-progress runs look like transient buffering rather than failures; the "Analysis skipped" warning is suppressed while a run is still in flight and only surfaces after the run reaches a terminal state; and Cancel now reliably targets the correct run by its unique ID rather than the short automation ID.

Features

• Auto-navigate to citations on card expand: Expanding a review checklist prompt card now automatically scrolls the PDF to the card's first citation — flagged findings are prioritized, and the lowest page number is selected first when multiple citations exist.
• Single-citation highlight: Only one citation is highlighted in the PDF at a time. Clicking a different finding row moves the highlight to that citation, making it easier to trace individual findings without visual clutter.
• Redesigned FSR export cover sheet: The annotated PDF export cover sheet has a new layout and typography (IBM Plex Mono tick marks), clickable rows that link directly to relevant findings, and now includes the Review Summary section.

Improvements

• Foot/Crossfoot action buttons always visible: Save, Reset, and Apply are now rendered inline at the bottom of the calculator card with an independently scrollable operand list — the action row stays in view regardless of how long your operand list is.

Bug Fixes

• Workpaper saves target the correct task: Fixed an issue where a queued autosave could be sent to the wrong task when navigating quickly between tasks while the memo panel stayed mounted.
• PDF navigation for bbox-less citations: Clicking a citation that has a page number but no drawn bounding box now correctly navigates the PDF to that page.
• Cross-page operands persist on save: Fixed an issue where cross-page Foot/Crossfoot operands were dropped when using the standalone save action.

File Explorer

• Multi-file selection across datasources: checkboxes appear on hover; a header checkbox selects every file across all datasource folders at once; Shift+click range-selects including across folder boundaries; folder headers show a live selected/total count (e.g., "3/8")
• Floating bulk action bar: when 2 or more files are selected, a pill-style bar anchors to the bottom of the file explorer showing the selection count, a clear button, delete, and (for auditors) an Actions dropdown
• Bulk download: download all selected files as a single zip; files from multiple datasources are organized into labeled subdirectories; a single selected file downloads directly without zipping
• Bulk re-run OCR: re-run text recognition on all selected files at once from the Actions dropdown
• Bulk match grouping transitions: move selected files between match groupings — Unmatched, Match Unlocked, Match Locked, or Excluded; each option shows how many of the selected files are eligible
• Right-click bulk context menu: right-clicking any file in a multi-selection shows the bulk actions menu instead of the single-file menu
• Datasource folder collapse/expand: each datasource folder can be independently collapsed or expanded; newly uploaded datasources auto-expand; collapsed state survives data refreshes; an upload button appears on folder header hover (auditors only)

Deep Analysis

• Chat auto-opens after completion: when a deep analysis question finishes, the resulting chat opens automatically in the task chat sidebar — no need to navigate manually; freshly created chats appear without a page refresh
• Deep Analysis chats in history: Deep Analysis chats in the chat history sidebar display a purple bolt icon to distinguish them from user-started conversations; the "Deep Analysis: " title prefix is stripped so only the question title is shown
• Faster start: deep analysis no longer waits for embedding or classification pipelines — questions begin processing immediately after OCR completes

Financial Statement Review

• AI math correction: the FSR pipeline now includes automatic math validation correction — for each detected table, the AI identifies total rows and columns, determines the correct operand set and sign assignments (add vs. subtract), and commits corrections only when the result achieves zero discrepancies; partial improvements are never applied; discovery mode also establishes operand formulas for totals that were detected by OCR but had no formula yet, improving validation coverage after the first AI run without requiring a separate follow-up pass

Workpaper

• Table row extraction completeness: the AI extraction safety net now catches leading-row omissions in financial statement tables (e.g., combined net assets or partners' capital statements with no parent header row); both leading-row gaps and zero-row returns trigger a correction re-prompt, ensuring all values appear in the workpaper grid
• Reviewed cells stay highlighted during Run All: accepted (green) and rejected (red) cells no longer flicker white or yellow mid-run when bulk automation is triggered — their visual state is maintained throughout the entire run; as an added safeguard, reviewed cells are now skipped even if the protected flag is absent

Engagement Board

• Collapsible document search results: document search results are now organized as collapsible groups — task → datasource → file → matching pages; each level is independently collapsible; tasks that matched only on title or description appear at the top labeled "Metadata match" with no document evidence shown; clicking a task header navigates to the task overview; clicking a file result opens it at the matching page

Co-Audit Agent

• Column scope for run_automation: the AI can now re-run a single attribute column across all rows in a datasource — the right action after updating a column's prompt and wanting every row recomputed without triggering a full workpaper re-run

Clients

• Filter banner: when active filters hide one or more items from the client list, a banner appears showing the hidden count with a Clear Filters button to reset all active filters at once

New Features

• Unmatched aggregate row: Match-agent datasources now display a pinned Unmatched row at the bottom of the workpaper grid whenever files remain unmatched. Each datasource column shows an amber chip with the unmatched file count — clicking it opens the cell detail panel filtered to those files. Once all files are matched, the chip turns green. Sample datasource columns are visually subdued so the row reads as a single status band rather than live data.

• Lock / Unlock Matches: A new Lock Matches button in the unmatched-files panel lets auditors bulk-protect every confirmed match in a datasource with a single click. When all matches are already protected, the button switches to Unlock Matches. The button is hidden while unmatched files remain — a Re-run Match button is shown instead.

Improvements

• Accounting Policy entries previously labeled Rich Text are now labeled Custom.

New

• Automatic client-upload enrollment: document group datasources added to a task that is already client-visible are now automatically set up for client uploads at creation time — no need to manually toggle "Provided by: Client" afterward.
• Co-Audit Agent row pagination: the agent can now page through large datasource tables (default 20 rows per call, up to 100 per request with offset support) and filter by specific columns — reducing incomplete or truncated reads on wide or high-row-count tables.

Improvements

• Unsupported file formats now download automatically: clicking a file in the file explorer that cannot be previewed triggers a browser download instead of failing silently.
• Clear message for deleted files: if a file has been removed, the viewer now shows "This file is no longer available — it may have been deleted." instead of a generic error.

Fixes

• Memo auto-saves are more resilient — edge cases that previously caused silent save failures are now handled correctly.
• Resolving multiple comments at once is faster — previously sent one request per comment, now resolved in a single call.
• Stale chat sessions are cleaned up correctly when a chat no longer exists in the list.

New Features

• Template share links: Copy a universal link to any task template directly from the template preview header. Recipients are taken to their own board with the Add modal already open to that template's focused preview — no searching required.
• Template deep-link: Boards opened with a ?templateId URL (e.g. from a shared template link) jump straight to that template's focused preview; an error toast appears if the template is unavailable or inaccessible.

Improvements

• Roll Forward now preserves more: rolling tasks into a new engagement carries over categories, memo fields (Purpose, Source, Scope, Procedures), and datasource column structure. Task status resets to To Do; conclusions are intentionally left blank. Engagement variables are cloned with their workpaper column links intact; re-running roll-forward removes any stale column links that were disconnected in the source since the last run.
• Add task modal context: when opened from an engagement, the client and engagement are now pre-selected from the current URL even before the page context has fully loaded. Client and engagement dropdowns scroll when the list is long.
• Task description editing: external changes to a task description (e.g. from AI chat) now apply in-place without reloading the editor — cursor position and focus are preserved mid-keystroke. Saving your own changes no longer briefly echoes back to the same tab.
• Formula columns on Collective output: formula columns that reference a Collective retrieval column now reduce the multi-value result sensibly — numbers are summed, dates become a min–max range, booleans check that all values are truthy, and text values are deduplicated and joined.
• Semantic search on Excel and CSV files: files without vector embeddings (Excel, CSV) now automatically fall back to trigram keyword search so they appear in semantic search results even before the embedding pipeline has run.
• Sandbox reset reliability: resetting a sandbox engagement now performs an atomic swap — the new copy replaces the old one with the same short identifier preserved. If the clone fails, the original is left intact and the reset can be retried. The board auto-refreshes on completion.

Bug Fixes

• Engagement access errors: navigating to an engagement you cannot access or that no longer exists now shows a descriptive error page instead of an infinite loading spinner. Auditors see guidance to ask an admin to add them; clients are told to contact their auditor. A Go to home button is shown in both cases.
• Clients page access: non-auditor users are now correctly blocked from /clients routes at the middleware layer and redirected to home, rather than reaching an error state.

New Features

• Memo fields in task templates: Templates now capture the task memo — Purpose, Source, Scope, and Procedures — when saved. When you create a task from a template, those fields are pre-filled automatically. The template browser also shows a read-only Memo preview before you select, so you know what setup content will be applied. Conclusion is intentionally excluded (it's per-audit sign-off content, not template material).
• Full-screen task chat: A new expand button in the task chat sidebar header widens the chat panel to fill the screen, giving more room for longer AI-assisted review sessions. Press ESC or click the collapse button to return to the normal layout.

Improvements

• Workflow builder retired: The inline workflow builder has been removed from Co-Audit Chat. Workpapers are now created directly via task creation. Existing conversations that used workflow builder tools display an inline notice in place of the old tool cards.

Bug Fixes

• Comment errors now visible: Comment submissions that failed silently (403 error) now show an error toast so you know the comment didn't go through.
• Bulk-review cross-datasource cells: Fixed bulk-review and right-click review not correctly picking up cross-datasource and top-level cells.
• Deep analysis progress indicator: The in-progress spinner now appears correctly when running a single-question deep analysis.
• Duplicate task preserves Conclusion: Duplicating a task now correctly retains the Conclusion memo field on the copy.

Workpaper

• Bulk accept/reject now covers all row types: Validate, Flag, and Clear operations in the range selection panel now apply to consolidated cells (cross-datasource inline rows) and top-level non-group rows — these were previously silently excluded. The accepted/rejected/pending counts in the range panel also update in real-time as review states change from any surface (right-click menu, live sync), without requiring you to re-select the range.
• Deep analysis spinner appears immediately: Starting a deep analysis run — whether a single question or a bulk run — now shows the progress spinner right away in the toast notification, instead of briefly displaying an empty circle before the run is registered server-side.
• Crash fixes in the workpaper editor: Two stability issues were resolved: one that could crash the workpaper when collapsing the chat panel via ESC in certain edge cases, and one triggered by mention transforms in the text editor.

Chat

• PDF-to-Excel exports work in follow-up messages: You can now ask the AI to export a PDF to Excel in a follow-up message, even after the original attachment has scrolled out of the latest turn. Previously the export tool was only available in the same message where the file was attached, making common workflows like "now export that to Excel" fail unexpectedly.
• Reorganized task chat sidebar header: Navigation controls (back to board, chat history) and action controls (new chat, expand/collapse, more options) are now split into distinct groups in the task chat header, improving discoverability and reducing visual clutter.

Formulas

• More reliable collective formula reductions: Formulas that aggregate multiple AI-retrieved values now handle mixed types correctly — numeric comparisons, boolean logic, and currency symbols (including $ prefixes) are preserved across collective reductions in AI retrieval formula paths.

Workpaper Grid

• Simplified row layout: single-child lineage paths now collapse inline — when a parent record has exactly one matched child, that child's values appear directly on the parent row, eliminating redundant rows. Multi-child fan-out continues to show individual child rows.
• Always-expanded rows: parent rows are no longer collapsible. Child rows appear automatically when there is genuine fan-out (2 or more children in the same datasource), removing the need to manually expand rows to see data.
• File pills on match columns: MATCH and MATCH_AGENT columns now display matched files as styled pills with a file-type icon and truncated filename. Hover to see the full name. Unmatched records show a red ✗. Contra-evidence count appears as a red badge on the pill corner.

Memo

• Editable memo fields: Purpose, Source, Scope, Procedures, and Conclusion are now editable directly on the memo page and persist on save.
• Editable testing attribute descriptions: testing attribute descriptions on the memo page can be edited inline, with hover controls to delete or link entries.
• Memo included in exports: memo fields now flow through the engagement-level export.

Automation

• Unassigned files excluded from Run All: records in match-result datasources with no parent assignment (files the match agent could not link upstream) are excluded from Run All automation passes. These records have no lineage context and previously produced unreliable downstream results.
• Lineage scoping tightened: downstream MATCH datasources with no parent records in the selected lineage scope are now fully skipped, preventing unintended record recreation during lineage runs.

Cell Detail Panel

• Inline citation pills: citation references in AI reasoning are now rendered as interactive pills rather than raw tags, including bare [N] tags the model occasionally produces.
• Match actions: the panel now wraps match cells in an Answer section with Re-run and Unlock actions directly accessible.
• Inline editing commits on blur: editors in the cell detail panel now save automatically on blur — no explicit save/cancel step required.

Bug Fixes

• Activity feed client visibility: auditor-only TEXT fields are no longer shown to clients in the activity feed or email notifications.
• Memo field updates: fixed an issue where memo field updates and their activity log entries were silently failing.
• Sidebar width: fixed a crash caused by invalid persisted sidebar width values.

AI & Chat

• Post-run automation summary: After triggering a workpaper automation from chat, the AI now automatically follows up with a ranked summary of proposed fix groups — most impactful changes first, with representative before/after examples — and asks for your authorization to apply them. If the run failed, it explains what went wrong. This summary appears immediately after the run and persists on subsequent chat revisits.
• Engagement search when creating tasks: When asking the AI to create a task from chat, you can now refer to an engagement by name, client, or short ID (e.g. "QA agentive") and the AI will filter to the right match. If nothing matches, you'll get a helpful suggestion rather than an error.

Workpaper

• Match source markers: Cells in match agent columns now display a small superscript on the passing icon showing how each protected record was sourced — ᶜ for client-assigned records, ᴹ for manually protected or matched records, and no suffix for agent-auto-matched records. These markers are preserved when exporting to Excel.
• Lasso deduplication: Manually lassoing a PDF region for a record that already exists (same file, same parent) now updates and promotes that row in place rather than inserting a duplicate row.
• Manual match acceptance deduplication: Manually accepting a file or record match in a match-agent datasource now promotes any existing agent-created rows for the same file/parent combination rather than creating duplicates — ensuring your manual acceptance survives the next agent cleanup pass.

Features

• Co-Audit Chat — The AI assistant has been renamed from "Copilot" to Co-Audit Agent across the platform, reflecting its deeper integration into the audit workflow.
• In-chat navigation — Ask the Co-Audit Agent to take you anywhere on the platform and it will present a confirmation card ("Navigate to [page]?") before moving you there. Your chat history carries forward when navigating to pages with a chat area (workpaper, overview, request table, task chat), so you never lose context mid-conversation.
• Task search and discovery from chat — The Co-Audit Agent can now search for tasks by name, description, or short identifier across all your engagements, and list all engagements you have access to — directly from the chat input.
• Create tasks from chat — Ask the agent to create a new task and it will do so without leaving the conversation. You can create blank tasks or instantiate from a firm template (pre-configured with datasources, columns, and automations). A "Task created" card appears with Go to task / Stay here options, and your current chat carries forward to the new task.
• Acceptance criteria tools in workpaper chat — The Co-Audit Agent can now list, create, edit, delete, and run acceptance criteria on a workpaper datasource directly from the task chat. Deletion requires explicit user approval. The agent will not proactively create acceptance criteria — it only acts when you ask.

Improvements

• File search is more powerful — searchFilePages now supports regex patterns, page ranges, case-sensitive mode, and word/line-based context extraction. It also auto-detects total page count to prevent out-of-range searches, skips pages that error rather than treating them as matches, and returns a clean error immediately for inaccessible files.
• Deep analysis survives page refreshes — Run status (running / completed / error) is now persisted server-side. If you refresh the workpaper while a deep analysis is in progress, the spinner is restored immediately and the run continues uninterrupted.
• Deep analysis progress card stays in sync on reconnect — If your WebSocket reconnects mid-run, the deep analysis progress card is updated in place rather than being dropped or duplicated.

Improvements

• PDF to Excel on large documents: The extraction pipeline that cleans and fills converted tables now processes pages in batches. Large PDFs that previously failed mid-conversion due to size limits will complete successfully. Pages that are exceptionally large are escalated to a higher-capacity model; pages that still exceed limits are skipped gracefully rather than aborting the entire conversion.

Bug Fixes

• OCR retry reliability: If an OCR job had to retry after partially completing, subsequent attempts could fail with a conflict error — leaving the file stuck in a failed state. Retries now cleanly replace any partial results, so files recover correctly without manual intervention.

Fixes

• Table data extraction has been restored to consistent, accurate performance — auditors working with complex tables in uploaded documents will see reliable extraction results again.
• OCR document indexing has been improved to ensure page content within uploaded files is fully indexed, reducing gaps where content could be missed during AI-assisted review and search.

Copilot Chat

• Citation continuity across conversation turns: when the AI cites document evidence with numbered references like [1] or [2], those source links are now carried forward into follow-up exchanges — the AI can correctly refer back to previously cited content in multi-turn conversations without losing context
• Copy and download controls on AI tables: markdown tables in AI responses now include a toolbar to copy or download tabular data directly, eliminating the need for manual selection
• Built-in product coaching: ask the copilot where to find any platform feature and it will describe what it does and guide you to the right page — the full platform layout is now part of its context in every chat mode

Engagement Board

• Filters persist across engagement switches: search queries and active filter selections now carry over when switching between engagements in the same browser session; assignee and category filters are automatically reconciled against the new engagement's valid values, with unrecognized values quietly dropped

Review

• Calculate-First mode for foot/crossfoot validation: draw all operands on the PDF first, then apply the result to a cell — an alternative to the standard flow; footer action buttons (Save, Reset, Apply Valid/Exception) stay pinned and accessible while scrolling the calculator panel
• Assignment focus mode: when the system cannot automatically link a drawn operand to its row and column in the table, the calculator collapses and a focused card guides you step-by-step to draw the row and column headers in the PDF to complete the link; the calculator re-expands once the link is resolved

Task Overview

• Task properties update live from AI chat: status, priority, due date, assignee, and category now reflect changes made via AI chat in real time — no page refresh needed
• Description stays current without refresh: the task description editor automatically remounts when AI chat updates the content externally, so edits appear immediately

Workpaper

• OCR is resilient to server interruptions: OCR processing now runs as a durable background job — if the server restarts mid-upload, the job is automatically picked up and completed on another node; duplicate OCR requests for the same file within 5 minutes are deduplicated; files no longer get stuck in a failed state due to transient server issues
• Formula cells export as live Excel formulas: cells whose value starts with = are written as real Excel formulas in all .xlsx exports — values calculate automatically when the file opens; number formatting (currency, date, number) is applied based on the column's data type
• Testing attribute columns are visually distinct: columns used for testing procedures now display with a blue header background and bold centered label, always appearing as the rightmost columns within their datasource group for quick identification
• Automation output reliability improved: AI-generated text in match and testing attribute automations is sanitized before saving, preventing rare write failures caused by edge-case characters in LLM outputs

New Features

• Export chat to Word: Download the full conversation from any task chat as a .docx file — available in the three-dot menu in the task chat sidebar.
• Engagement disambiguation: When multiple engagements share the same short identifier (e.g. ACME-2025), navigating to that engagement now shows a disambiguation screen listing each match with its client name and audit period. Open the right engagement directly or jump to its settings from the same screen.

Improvements

• AI analysis performance: Workpaper AI runs now process document pages more efficiently, reducing wait times for large or complex files.
• FSR year detection: Tables with ambiguous year references are classified more accurately, reducing mismatches during financial statement review.

Bug Fixes

• Chat overflow: Long URLs and text in chat messages now wrap correctly instead of overflowing or being clipped.
• Requests with files counter: The "Requests with files" count in the request table now stays accurate after uploading files individually rather than in bulk.

• Faster search results — Document search results are now available sooner after file upload, with more precise page-level matches.
• More reliable automations — Multi-step automation workflows are less likely to fail silently, especially on engagements with many workpapers running concurrently.
• Improved name matching — Automations are better at identifying the same entity across different documents, reducing false mismatches.
• Fewer processing failures — Documents with unusual formatting or large text blocks are less likely to fail during upload processing.

New Features

• Word document export from Copilot: Ask Copilot to generate a Word (.docx) file — it will produce a formatted document from your conversation and deliver it as a download button directly in the chat. Supports headings, lists, and tables.

Improvements

• Redesigned board export: The export button now opens a modal with two distinct sections:
  • Status Report — export tasks to Excel filtered by one or more statuses. Client users see requests labeled accordingly and cannot include backlog items.
  • Files + Workpapers (auditors only) — choose your format: a .zip archive compatible with the Citations Excel Add-in, a single .xlsx with embedded files, or a workpapers-only .xlsx. Workpaper sheets now include threaded Excel comments drawn from task activity at the cell, row, and file level.
• Cleaner export experience in chat: After Copilot generates an Excel or Word file, a download button appears in the chat — no stray or broken download links.