Merge branch 'main' into improve_document_analysis

app/MindWork AI Studio/wwwroot/app.css

@@ -475,20 +475,59 @@ tr:has(> .provider-group-header) .mud-icon-button {
  * Rows of the tool selection which the chat and the assistants open from their footer. There will be
  * far more tools than the ones we start with, so a row must not waste height: MudBlazor's settings
  * button alone puts 12px of padding around a 24px icon, which makes a row 48px tall before the
- * switch and the frame are counted at all. Size.Small takes most of that away; the rule below takes
- * the rest, and it has to name the MudBlazor class to outweigh its specificity. Alternating rows
- * carry a grey ground, which tells a long list apart better than a separator line does and costs no
- * height at all. The colors are MudBlazor palette variables, so both grounds follow the theme.
+ * switch and the frame are counted at all. Size.Small takes most of that away; the rules below take
+ * the rest, and the second one has to name the MudBlazor class to outweigh its specificity.
  */
 .tool-selection-rows > .tool-selection-row {
     padding: 0.15rem 0.25rem;
-    border-radius: var(--mud-default-borderradius);
-}
-
-.tool-selection-rows > .tool-selection-row:nth-child(odd) {
-    background-color: var(--mud-palette-background-gray);
 }
 
 .tool-selection-row .mud-icon-button {
     padding: 0.2em;
 }
+
+/*
+ * Rows of the data source lists, in the popover next to the tool selection as well as in the
+ * settings dialog. A row carries a name and at most one icon, so there is no reason for it to be
+ * 48px tall: MudBlazor pads the item with 8px on both sides and the text slot with another 4px,
+ * which is more frame than content. Dense on the list halves the first part, the rule below takes
+ * the second one away, and it has to name the MudBlazor class to outweigh its specificity.
+ */
+.data-source-rows .mud-list-item-text {
+    margin-top: 0;
+    margin-bottom: 0;
+}
+
+/*
+ * The checkboxes MudBlazor renders into a multi-selection list come out larger than the box of the
+ * tool selection next to it, and MudList has no parameter for their size. So the three rules below
+ * state it: the 20px icon and the 4px of padding which Size.Small together with Dense produce over
+ * there, plus the 4px between the box and the name which the tool row takes from the spacing of its
+ * stack -- the list puts its checkbox outside the slot that holds our own markup, so no stack of
+ * ours reaches it. The icon needs a rule of its own because MudBlazor gives it an explicit font
+ * size, which no inherited one can outrank.
+ */
+.data-source-rows .mud-checkbox {
+    margin-inline-end: 0.25rem;
+}
+
+.data-source-rows .mud-checkbox .mud-icon-button {
+    padding: 0.25rem;
+}
+
+.data-source-rows .mud-checkbox .mud-icon-root {
+    font-size: 1.25rem;
+}
+
+/*
+ * The frame around a text switch in its compact form. MudBlazor pads the slot of an outlined field
+ * with 18.5px above and below, which is the right amount for the line of text such a field usually
+ * holds -- a switch of 24px is left swimming in the middle of it. Size.Small already took the switch
+ * down; this brings the frame with it, and it has to name the MudBlazor classes to outweigh their
+ * specificity. Only the two vertical values of that shorthand are replaced, so the 14px to the left
+ * and right stay as they are.
+ */
+.text-switch-dense .mud-input-slot.mud-input-root-outlined {
+    padding-top: 0.5rem;
+    padding-bottom: 0.5rem;
+}

app/MindWork AI Studio/wwwroot/changelog/v26.9.1.md

@@ -11,12 +11,16 @@
 - Added support for OpenAI's GPT-6 Astra.
 - Added the context window to what AI Studio knows about a model, wherever its metadata states one.
 - Added a live read of that context window at the providers which report it, among them Mistral, Groq, OpenRouter, and self-hosted vLLM servers. You then get the window your own server was started with, not the one the model card advertises.
-- Added a token count below the message field, so you always see how much of the conversation you have used. It can be an estimate when a provider does not give AI Studio everything it needs to count exactly.
+- Added a token count below the message field, so you always see how much of the conversation you have used. It counts everything that travels along: your messages, the files you attached, what your data sources contributed, and the tools you offered the AI. It can be an estimate when a provider does not give AI Studio everything it needs to count exactly.
 - Added a warning when your conversation holds more images than the model accepts, wherever we know that limit. The Visual Briefing assistant stops before anything is uploaded, instead of letting the provider refuse it afterward.
 - Added the context window and the image limits to the expert provider settings, next to the abilities you could already state there. Leave a field empty, and AI Studio keeps its own answer, which you see as the placeholder. IT departments can state the same numbers for the providers they roll out.
 - Added model plugins, so IT departments can describe the models their organization runs itself.
 - Added local RAG as a beta feature, so the AI can answer from your own documents. You point AI Studio at a folder or at a single file, and it prepares those documents in the background so their contents can be found again later. Ask a question with such a data source selected, and AI Studio looks for the passages that fit your question and hands only those to the model, along with where each one came from. We will keep developing it together with the people who use it: to try it, open the app settings, allow preview features down to beta, and then enable the RAG feature. Many thanks to Paul Koudelka (`PaulKoudelka`) for around ten months of work on the concept and the implementation.
 - Added the setup for local data sources. You pick an embedding provider, and AI Studio asks for your confirmation before any document goes to a cloud service. It keeps up with your files as they change, shows the progress on a page of its own, and checks every document for hidden instructions before indexing it. Documents without readable text, such as scanned pages, are remembered as such, so AI Studio does not work through them again after every start — it comes back to them once they change.
+- Added a way to open the sources of your own documents: click a source below an answer, and the document opens in the program your system uses for it.
+- Added a jump to the right page for the sources of your own documents (RAG), so a PDF opens directly where the passage was found, wherever your system and its program support it.
+- Added a way to show a source of your own documents (RAG) in your file manager.
+- Added the page of a passage to what the AI is told when it answers from your own documents (RAG), so it can name the page an answer rests on.
 - Added support for several drop areas on the same page. More complex assistants can now receive files or folders by drag and drop at more than one place.
 - Added drag and drop to the input and output folder of the Batch Processing assistant: drop a folder onto either field to choose it.
 - Added ways to load text from a file and drop zones for them, throughout the assistants and dialogs. We went through them one by one, so many fields that used to accept typed text only now take the content of a file as well.
@@ -42,7 +46,9 @@
 - Fixed the copy button leaving the sources behind. Copy an answer, and its sources come along.
 - Fixed a tile that opens a chat directly always demanding a workspace. Leave the workspace empty in the Assistant Builder, and the tile opens a disappearing chat instead.
 - Fixed the same restriction for plugin authors: a direct-chat launcher can now open a chat without naming a workspace. The example assistant plugin shows both ways.
-- Fixed an answer built without your data sources looking exactly like one that used them. When AI Studio cannot reach the sources you picked, it now tells you instead of quietly answering without them.
-- Fixed the same silence when the step that picks the fitting passages out of your documents cannot run. You are told that the answer rests on everything that was found.
+- Fixed data sources you picked for your chats vanishing from the selection without a word when they cannot be used. AI Studio now lists them by name, so you can see why an answer was created without them.
+- Fixed the data sources you picked for a chat being forgotten the moment you changed your selection while one of them could not be used. Such a source stays selected and is used again as soon as it is available.
+- Fixed the silence when the step that picks the fitting passages out of your documents fails. You are told that the answer rests on everything that was found.
 - Fixed the regenerate button taking an answer away without producing a new one. This happened in chats started from a template that holds no question of your own.
+- Fixed the counter above an answer, which shows how many sources it rests on, doing nothing when you clicked it. It now takes you down to the sources.
 - Upgraded the Visual Briefing Assistant (in preview) from the prototype to the beta state. The assistant is now completely implemented and is undergoing a deeper testing phase in preparation for release. To try it, open the app settings, allow preview features down to beta, and then enable the Visual Briefing Assistant there.