Changelog

Adds SpellCheckAdorner (wavy underlines for misspelled words in TextBox),

SpellCheckBehavior (attached behavior to enable spell check), autocorrect

service, and spell check rendering utilities. Integrates spell check into

RichTextEditor with debounced recheck and dictionary change handling.

Fixes Timer ambiguity (System.Timers vs System.Threading) in SpellCheckAdorner

and RichTextEditor.SpellCheck by fully qualifying System.Timers.Timer. Fixes

IsAttachedToVisualTree → GetVisualRoot() and GetTemplateChildren →

GetVisualDescendants for Avalonia 11.3.x compatibility.

The Notes RAG indexer was looking for a top-level "blocks" array on page

entities, but pages store content as content: { type: "doc", content:

[...blocks...] } (the PageDocument structure). This meant only page

titles were indexed — no actual block content reached the RAG vector

store. Fixed ExtractPageChunks to navigate content.content for the

blocks array, with legacy "blocks" fallback. Also extended

ExtractBlockText to handle bullet/numbered/task list items (text is

in items[].text) and nested children blocks.

Enriched the cloud system prompt with personality traits: honest but

empathetic tone, balanced verbosity, clean dry humor, neutral error

handling that pivots to solutions. Technical depth now adapts to the

user via memory context rather than assuming expertise level.

RAG search results were dominated by the active plugin's data, making

cross-plugin queries (e.g. asking about Notes from Tasks) return no

results from the mentioned plugin. Fix: fetch 3x candidates, then

apply plugin diversity via round-robin interleaving. When the user

mentions a specific plugin by name (e.g. "notes", "calendar"),

half the result slots are reserved for that plugin's data.

The root cause of the AI responding conversationally ("Shall I go ahead?"

"Creating those now!") without actually executing actions was that RAG

returned 0 intent_action chunks. Entity data (the user's actual tasks,

notes, etc.) crowded out intent_action chunks in the general semantic

search because task titles are closer to the query than "ACTION: Create

Task" descriptors.

Fix: BuildRagContextWithIntentsAsync now performs TWO parallel searches —

one general (for entity data + plugin context) and one specifically

filtered to intent_action entity type. Results are deduped and merged,

ensuring the AI always sees relevant available actions alongside the

user's data context.

Also fixed incorrect slot names in ActionFormatHeader for tasks.add_link

(was source_task_id/target_entity_id, corrected to task_id/target_id).

Desktop version 1.65.6 → 1.65.7.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>