Merge remote-tracking branch 'gitea/develop' into fix/522-internal-links

Reviewer follow-up on the #522 internal-link import fix. The fix itself is
confirmed correct; these three changes harden its load-bearing subset invariant
and fix a stale doc.

Finding 1 (drift guard): the in-package test only compared isInternalPagePath
against a HAND-COPIED server regex, so a later narrowing of the real server
INTERNAL_LINK_REGEX would leave the test green while the client silently became
WIDER than the server. Add apps/server/.../export/internal-link-parity.spec.ts:
the top layer already depends on @docmost/prosemirror-markdown and exports
isInternalPagePath, so this spec imports the LIVE server INTERNAL_LINK_REGEX AND
the LIVE isInternalPagePath and asserts subset over an accept-corpus — a server
narrowing now reddens CI. The in-package manual copy is demoted from its
"drift guard" role (comment updated to say so; it stays as local documentation).

Finding 2 (charset non-vacuity): the REJECT list was purely structural, so the
mutation widening the slug charset [a-zA-Z0-9-] -> [a-zA-Z0-9-.] survived the
whole suite. Add REJECT cases whose ONLY defect is a forbidden slug character
(abc.def / abc_def / abc%20 / abc~x); assert both isInternalPagePath and the
server regex reject them. The mutation now reddens the new reject test.

Finding 3 (stale JSDoc): canonicalize.ts KNOWN_DEFAULTS module blurb claimed
every entry was read from docmost-schema and import-materialized. The new
link.internal default breaks both claims (source is editor-ext/src/lib/link.ts;
external links leave internal absent/null, never false). Add a bullet documenting
the one editor-sourced, non-materialized default (false ≡ absent/null).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

CHANGELOG.md

@@ -129,6 +129,13 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### Added
 
+- **A drifted comment suggestion can be re-synced instead of failing forever
+  with a 409.** A suggestion whose stored anchor no longer matched the live
+  document used to reject every apply attempt with an unrecoverable conflict; a
+  new resync path re-reads the live anchor so the suggestion applies against the
+  current text, and orphaned anchors (whose marked run was deleted) are
+  reconciled rather than left blocking. (#496)
+
 - **Place several images side by side in a row.** A new "Inline (side by
   side)" alignment mode in the image bubble menu renders consecutive inline
   images as a row that wraps onto the next line on narrow screens. The row is
@@ -352,7 +359,20 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
   body-timeout so a legitimate >1-min idle between the model's tool calls no
   longer breaks a long-lived SSE socket (new `AI_MCP_SSE_BODY_TIMEOUT_MS`, default
   10 min; see `.env.example`). (#489)
-
+- **Decisions on comment suggestions now leave a durable audit record.**
+  Applying or dismissing a comment suggestion hard-deletes the (childless)
+  subject comment, so the only surviving trace of who decided what is the audit
+  event — but the audit trail was wired to a Noop service that silently
+  swallowed every event. The trail is now DB-backed, so
+  `comment.suggestion_applied` / `comment.suggestion_dismissed` (and the other
+  comment-decision events) persist to the `audit` table and can be reviewed
+  after the comment is gone. A persistence failure is still swallowed with a
+  warning so it never breaks the originating request. (#496)
+- **Applying a comment suggestion no longer strips the replaced run's inline
+  formatting.** The suggested text was re-inserted carrying only the comment
+  anchor mark, silently dropping bold/italic/code/link on the affected run; the
+  prevailing formatting of the replaced run is now carried onto the applied
+  text. (#496)
 - **The server no longer runs out of heap during long autonomous agent runs.** A
   new pnpm patch on `ai@6.0.134` stops the SDK from building a cumulative
   snapshot of the ENTIRE turn text on every streamed text-delta when no output

apps/server/src/app.module.ts

@@ -25,7 +25,7 @@ import { CacheModule } from '@nestjs/cache-manager';
 import KeyvRedis from '@keyv/redis';
 import { LoggerModule } from './common/logger/logger.module';
 import { ClsModule } from 'nestjs-cls';
-import { NoopAuditModule } from './integrations/audit/audit.module';
+import { AuditModule } from './integrations/audit/audit.module';
 import { ThrottleModule } from './integrations/throttle/throttle.module';
 import { McpModule } from './integrations/mcp/mcp.module';
 import { SandboxModule } from './integrations/sandbox/sandbox.module';
@@ -55,7 +55,7 @@ try {
       middleware: { mount: true },
     }),
     LoggerModule,
-    NoopAuditModule,
+    AuditModule,
     CoreModule,
     DatabaseModule,
     EnvironmentModule,

apps/server/src/collaboration/yjs.util.spec.ts

@@ -529,4 +529,107 @@ describe('replaceYjsMarkedText', () => {
     expect(result).toEqual({ applied: false, currentText: 'abcdef' });
     expect(text.toDelta()).toEqual(before);
   });
+
+  // #496: apply must NOT silently strip the replaced run's inline formatting.
+  // Build a paragraph and format the marked range with extra marks, then assert
+  // the replacement carries them.
+  function buildFormatted(
+    runs: Array<{ text: string; attrs?: Record<string, any> }>,
+  ): { fragment: Y.XmlFragment; text: Y.XmlText } {
+    const ydoc = new Y.Doc();
+    const fragment = ydoc.getXmlFragment('default');
+    const para = new Y.XmlElement('paragraph');
+    fragment.insert(0, [para]);
+    const text = new Y.XmlText();
+    para.insert(0, [text]);
+    text.insert(0, runs.map((r) => r.text).join(''));
+    let offset = 0;
+    for (const run of runs) {
+      if (run.attrs) text.format(offset, run.text.length, run.attrs);
+      offset += run.text.length;
+    }
+    return { fragment, text };
+  }
+
+  it('preserves the original run formatting (bold + link) on the replacement', () => {
+    const { fragment, text } = buildFormatted([
+      { text: 'see ' },
+      {
+        text: 'old',
+        attrs: {
+          comment: { commentId: 'c1', resolved: false },
+          bold: true,
+          link: { href: 'https://x.test' },
+        },
+      },
+      { text: ' end' },
+    ]);
+
+    const result = replaceYjsMarkedText(fragment, 'c1', 'old', 'new');
+
+    expect(result).toEqual({ applied: true, currentText: 'new' });
+    // The comment anchor AND the bold/link marks survive the delete+insert.
+    expect(text.toDelta()).toEqual([
+      { insert: 'see ' },
+      {
+        insert: 'new',
+        attributes: {
+          comment: { commentId: 'c1', resolved: false },
+          bold: true,
+          link: { href: 'https://x.test' },
+        },
+      },
+      { insert: ' end' },
+    ]);
+  });
+
+  it('mixed formatting under the mark: replacement takes the DOMINANT (longest) run, NOT the leading one', () => {
+    // Leading run is SHORT + plain ("x", 1 char); the following run is LONGER +
+    // bold ("bolded", 6 chars), same commentId. The longest run is deliberately
+    // NOT first: a "first-wins" pick would carry plain (no bold), so asserting
+    // bold on the result only holds if the code genuinely selects the LONGEST run.
+    const { fragment, text } = buildFormatted([
+      { text: 'x', attrs: { comment: { commentId: 'c1', resolved: false } } },
+      {
+        text: 'bolded',
+        attrs: { comment: { commentId: 'c1', resolved: false }, bold: true },
+      },
+    ]);
+
+    const result = replaceYjsMarkedText(fragment, 'c1', 'xbolded', 'Z');
+
+    expect(result).toEqual({ applied: true, currentText: 'Z' });
+    expect(text.toDelta()).toEqual([
+      {
+        insert: 'Z',
+        attributes: { comment: { commentId: 'c1', resolved: false }, bold: true },
+      },
+    ]);
+  });
+
+  it('mixed formatting under the mark: on a length tie the FIRST run wins', () => {
+    // Two equal-length runs (2 chars each) with different formatting, same
+    // commentId. The reduce keeps the accumulator on a tie, so the FIRST run
+    // (italic) prevails over the later bold one.
+    const { fragment, text } = buildFormatted([
+      {
+        text: 'AA',
+        attrs: { comment: { commentId: 'c1', resolved: false }, italic: true },
+      },
+      {
+        text: 'BB',
+        attrs: { comment: { commentId: 'c1', resolved: false }, bold: true },
+      },
+    ]);
+
+    const result = replaceYjsMarkedText(fragment, 'c1', 'AABB', 'Z');
+
+    expect(result).toEqual({ applied: true, currentText: 'Z' });
+    expect(text.toDelta()).toEqual([
+      {
+        insert: 'Z',
+        attributes: { comment: { commentId: 'c1', resolved: false }, italic: true },
+      },
+    ]);
+  });
 });

apps/server/src/collaboration/yjs.util.ts

@@ -145,6 +145,10 @@ type MarkedSegment = {
   length: number;
   text: string;
   markAttrs: Record<string, any>;
+  // The FULL attribute set of this delta run — the `comment` mark plus any
+  // inline formatting (bold/italic/code/link/…). Captured so apply can carry the
+  // original run's formatting onto the replacement instead of dropping it.
+  attributes: Record<string, any>;
 };
 
 /**
@@ -202,6 +206,7 @@ export function replaceYjsMarkedText(
             length,
             text: insert,
             markAttrs: markAttr,
+            attributes,
           });
         }
         offset += length;
@@ -251,15 +256,25 @@ export function replaceYjsMarkedText(
     return { applied: false, currentText: joinedText };
   }
 
-  // 3. All guards passed: delete the marked run and re-insert newText with the
-  // same comment attributes at the same offset. Atomic within the caller's
-  // transaction.
+  // 3. All guards passed: delete the marked run and re-insert newText at the
+  // same offset. Atomic within the caller's transaction.
   const start = segments[0].offset;
   const len = segments.reduce((sum, s) => sum + s.length, 0);
-  const markAttrs = segments[0].markAttrs;
+
+  // Carry the ORIGINAL run's formatting onto the replacement (#496): inserting
+  // with only the `comment` mark silently dropped bold/italic/code/link of the
+  // replaced text. Yjs applies one flat attribute set to the whole insert, so
+  // when the marked run mixes formatting we pick the DOMINANT segment (the one
+  // covering the most characters) and apply its attributes — a v1 that preserves
+  // the common single-format case exactly and, for a mixed run, keeps the
+  // prevailing style rather than losing all of it. `attributes` already carries
+  // the `comment` mark (every collected segment is filtered on it above), so the
+  // anchor is preserved by copying the run's attribute set verbatim.
+  const dominant = segments.reduce((a, b) => (b.length > a.length ? b : a));
+  const insertAttrs = { ...dominant.attributes };
 
   node.delete(start, len);
-  node.insert(start, newText, { comment: markAttrs });
+  node.insert(start, newText, insertAttrs);
 
   return { applied: true, currentText: newText };
 }

apps/server/src/core/comment/comment.controller.ts

@@ -16,6 +16,7 @@ import { UpdateCommentDto } from './dto/update-comment.dto';
 import { ResolveCommentDto } from './dto/resolve-comment.dto';
 import { ApplySuggestionDto } from './dto/apply-suggestion.dto';
 import { DismissSuggestionDto } from './dto/dismiss-suggestion.dto';
+import { ResyncSuggestionAnchorDto } from './dto/resync-suggestion-anchor.dto';
 import { PageIdDto, CommentIdDto } from './dto/comments.input';
 import { AuthUser } from '../../common/decorators/auth-user.decorator';
 import { AuthWorkspace } from '../../common/decorators/auth-workspace.decorator';
@@ -235,6 +236,39 @@ export class CommentController {
     return this.commentService.applySuggestion(comment, user, provenance);
   }
 
+  @HttpCode(HttpStatus.OK)
+  @Post('resync-suggestion-anchor')
+  async resyncSuggestionAnchor(
+    @Body() dto: ResyncSuggestionAnchorDto,
+    @AuthUser() user: User,
+    @AuthWorkspace() workspace: Workspace,
+  ) {
+    const comment = await this.commentRepo.findById(dto.commentId, {
+      includeCreator: true,
+      includeResolvedBy: true,
+    });
+    if (!comment) {
+      throw new NotFoundException('Comment not found');
+    }
+
+    const page = await this.pageRepo.findById(comment.pageId);
+    if (!page || page.deletedAt) {
+      throw new NotFoundException('Page not found');
+    }
+
+    // Authorize BEFORE revealing structural detail (mirrors apply/dismiss).
+    // Re-anchoring does NOT change the page text — it only corrects the stored
+    // selection metadata — so the page-level gate is comment access. The service
+    // further restricts it to the suggestion's own author.
+    await this.pageAccessService.validateCanComment(page, user, workspace.id);
+
+    return this.commentService.resyncSuggestionAnchor(
+      comment,
+      dto.selection,
+      user,
+    );
+  }
+
   @HttpCode(HttpStatus.OK)
   @Post('dismiss-suggestion')
   async dismissSuggestion(

apps/server/src/core/comment/comment.service.apply-suggestion.spec.ts

@@ -146,11 +146,19 @@ describe('CommentService — applySuggestion', () => {
       'page-1',
       expect.objectContaining({ operation: 'commentDeleted', commentId: 'c-1' }),
     );
+    // #496: hard-deleted row → the audit payload is the only surviving record.
     expect(auditService.log).toHaveBeenCalledWith(
       expect.objectContaining({
         event: AuditEvent.COMMENT_SUGGESTION_APPLIED,
         resourceType: AuditResource.COMMENT,
         resourceId: 'c-1',
+        metadata: expect.objectContaining({
+          pageId: 'page-1',
+          suggestedText: 'new text',
+          selection: 'old text',
+          commentAuthor: 'user-1',
+          decidedBy: 'user-1',
+        }),
       }),
     );
     expect(result.outcome).toBe('deleted');
@@ -189,17 +197,25 @@ describe('CommentService — applySuggestion', () => {
     expect(resolvePatch.resolvedAt).toBeInstanceOf(Date);
     expect(resolvePatch.resolvedById).toBe('user-1');
 
-    // NOT deleted; broadcast an update, not a deletion.
+    // NOT deleted.
     expect(commentRepo.deleteComment).not.toHaveBeenCalled();
     expect(collaborationGateway.handleYjsEvent).not.toHaveBeenCalledWith(
       'deleteCommentMark',
       expect.anything(),
       expect.anything(),
     );
+    // #496 dedup: resolveComment broadcasts `commentResolved` with the enriched
+    // row; finalize must NOT ALSO emit a redundant `commentUpdated`. So the
+    // thread receives exactly ONE resolve broadcast and no update broadcast.
     expect(wsService.emitCommentEvent).toHaveBeenCalledWith(
       'space-1',
       'page-1',
-      expect.objectContaining({ operation: 'commentUpdated', comment: UPDATED }),
+      expect.objectContaining({ operation: 'commentResolved', comment: UPDATED }),
+    );
+    expect(wsService.emitCommentEvent).not.toHaveBeenCalledWith(
+      'space-1',
+      'page-1',
+      expect.objectContaining({ operation: 'commentUpdated' }),
     );
 
     expect(auditService.log).toHaveBeenCalledWith(
@@ -211,6 +227,36 @@ describe('CommentService — applySuggestion', () => {
     expect(result.outcome).toBe('resolved');
   });
 
+  it('re-entry: already applied+resolved WITH replies → emits commentUpdated (dedup does not over-suppress)', async () => {
+    // suggestionAppliedAt set → idempotent finalize; resolvedAt set → resolveComment
+    // is skipped, so there is NO commentResolved broadcast. The applied-stamp state
+    // must still reach clients via a single commentUpdated.
+    const { service, wsService } = makeService(
+      { applied: false, currentText: 'new text' },
+      true,
+    );
+
+    await service.applySuggestion(
+      suggestionComment({
+        suggestionAppliedAt: new Date(),
+        resolvedAt: new Date(),
+      }),
+      user(),
+    );
+
+    expect(wsService.emitCommentEvent).toHaveBeenCalledWith(
+      'space-1',
+      'page-1',
+      expect.objectContaining({ operation: 'commentUpdated', comment: UPDATED }),
+    );
+    // Nothing resolved this time (already resolved) → no resolve broadcast.
+    expect(wsService.emitCommentEvent).not.toHaveBeenCalledWith(
+      'space-1',
+      'page-1',
+      expect.objectContaining({ operation: 'commentResolved' }),
+    );
+  });
+
   // --- error / rejection branches -----------------------------------------
 
   it('applied=false and currentText differs → ConflictException with currentText in payload', async () => {

apps/server/src/core/comment/comment.service.dismiss-suggestion.spec.ts

@@ -107,11 +107,21 @@ describe('CommentService — dismissSuggestion', () => {
       'page-1',
       expect.objectContaining({ operation: 'commentDeleted', commentId: 'c-1' }),
     );
+    // #496: the row is hard-deleted, so the audit payload must carry the
+    // decision's substance (what was suggested, the anchored text, who authored
+    // it, who decided) — it is the only surviving record.
     expect(auditService.log).toHaveBeenCalledWith(
       expect.objectContaining({
         event: AuditEvent.COMMENT_SUGGESTION_DISMISSED,
         resourceType: AuditResource.COMMENT,
         resourceId: 'c-1',
+        metadata: expect.objectContaining({
+          pageId: 'page-1',
+          suggestedText: 'new text',
+          selection: 'old text',
+          commentAuthor: 'user-1',
+          decidedBy: 'user-1',
+        }),
       }),
     );
     expect(result.outcome).toBe('deleted');

apps/server/src/core/comment/comment.service.resync-anchor.spec.ts

@@ -0,0 +1,126 @@
+import { BadRequestException, ForbiddenException } from '@nestjs/common';
+import { CommentService } from './comment.service';
+
+/**
+ * Coverage for CommentService.resyncSuggestionAnchor (#496): re-anchoring a
+ * suggestion's stored selection (== apply-time expectedText) to the live-doc
+ * substring. The service is built directly with jest-mocked deps (the
+ * @InjectQueue tokens can't be resolved by Test.createTestingModule — see the
+ * sibling specs).
+ */
+describe('CommentService — resyncSuggestionAnchor', () => {
+  const UPDATED = { id: 'c-1', selection: 'new anchor', __updated: true } as any;
+
+  function makeService() {
+    const commentRepo: any = {
+      updateComment: jest.fn(async () => undefined),
+      findById: jest.fn(async () => UPDATED),
+    };
+    const service = new CommentService(
+      commentRepo,
+      {} as any,
+      { emitCommentEvent: jest.fn() } as any,
+      {} as any,
+      { add: jest.fn() } as any,
+      { add: jest.fn() } as any,
+      { log: jest.fn() } as any,
+    );
+    return { service, commentRepo };
+  }
+
+  const suggestion = (over?: Partial<any>): any => ({
+    id: 'c-1',
+    creatorId: 'user-1',
+    parentCommentId: null,
+    selection: 'old anchor',
+    suggestedText: 'new text',
+    suggestionAppliedAt: null,
+    resolvedAt: null,
+    ...over,
+  });
+  const user = (over?: Partial<any>): any => ({ id: 'user-1', ...over });
+
+  it('persists the new selection and returns the enriched comment', async () => {
+    const { service, commentRepo } = makeService();
+
+    const out = await service.resyncSuggestionAnchor(
+      suggestion(),
+      'new anchor',
+      user(),
+    );
+
+    expect(commentRepo.updateComment).toHaveBeenCalledWith(
+      { selection: 'new anchor' },
+      'c-1',
+    );
+    expect(out).toBe(UPDATED);
+  });
+
+  it('is idempotent: no write when the anchor already matches', async () => {
+    const { service, commentRepo } = makeService();
+
+    const out = await service.resyncSuggestionAnchor(
+      suggestion({ selection: 'same' }),
+      'same',
+      user(),
+    );
+
+    expect(commentRepo.updateComment).not.toHaveBeenCalled();
+    expect(out).toEqual(suggestion({ selection: 'same' }));
+  });
+
+  it('rejects a non-author (only the suggestion owner may re-anchor)', async () => {
+    const { service, commentRepo } = makeService();
+    await expect(
+      service.resyncSuggestionAnchor(suggestion(), 'new anchor', user({ id: 'other' })),
+    ).rejects.toBeInstanceOf(ForbiddenException);
+    expect(commentRepo.updateComment).not.toHaveBeenCalled();
+  });
+
+  it('rejects a reply / a comment with no suggestion', async () => {
+    const { service } = makeService();
+    await expect(
+      service.resyncSuggestionAnchor(
+        suggestion({ parentCommentId: 'p-1' }),
+        'x',
+        user(),
+      ),
+    ).rejects.toBeInstanceOf(BadRequestException);
+    await expect(
+      service.resyncSuggestionAnchor(
+        suggestion({ suggestedText: null }),
+        'x',
+        user(),
+      ),
+    ).rejects.toBeInstanceOf(BadRequestException);
+  });
+
+  it('rejects re-anchoring an already applied or resolved suggestion', async () => {
+    const { service } = makeService();
+    await expect(
+      service.resyncSuggestionAnchor(
+        suggestion({ suggestionAppliedAt: new Date() }),
+        'x',
+        user(),
+      ),
+    ).rejects.toBeInstanceOf(BadRequestException);
+    await expect(
+      service.resyncSuggestionAnchor(
+        suggestion({ resolvedAt: new Date() }),
+        'x',
+        user(),
+      ),
+    ).rejects.toBeInstanceOf(BadRequestException);
+  });
+
+  it('rejects a no-op selection equal to the suggested text', async () => {
+    const { service } = makeService();
+    await expect(
+      service.resyncSuggestionAnchor(
+        suggestion({ suggestedText: 'new text' }),
+        'new text',
+        user(),
+      ),
+    ).rejects.toBeInstanceOf(BadRequestException);
+  });
+});

apps/server/src/core/comment/comment.service.ts

@@ -370,6 +370,76 @@ export class CommentService {
     return updatedComment;
   }
 
+  /**
+   * Re-sync a suggestion's stored `selection` (== apply-time expectedText) to the
+   * RAW substring the inline mark actually covers in the LIVE document (#496).
+   *
+   * The MCP client creates the comment from a DEBOUNCED REST snapshot, then
+   * anchors the mark in the live collab doc. When the two disagree (the doc moved
+   * on in the debounce window) the stored selection no longer equals the marked
+   * text, so EVERY apply 409s ("the commented text changed"). After anchoring the
+   * client re-reads the exact marked substring and calls this to store it, making
+   * apply's strict equality hold.
+   *
+   * Only meaningful for an un-settled top-level suggestion authored by the
+   * caller: applying/resolving freezes the anchor, and a reply-carrying thread is
+   * preserved rather than mutated. The new text must still differ from the
+   * suggestion (else "apply" would be a no-op), preserving create()'s invariant.
+   */
+  async resyncSuggestionAnchor(
+    comment: Comment,
+    selection: string,
+    user: User,
+  ): Promise<Comment> {
+    if (comment.creatorId !== user.id) {
+      throw new ForbiddenException(
+        'You can only re-anchor your own suggestion',
+      );
+    }
+    if (comment.parentCommentId) {
+      throw new BadRequestException(
+        'Only a top-level comment can carry a suggested edit',
+      );
+    }
+    if (!comment.suggestedText) {
+      throw new BadRequestException('This comment has no suggested edit');
+    }
+    // A settled suggestion's anchor is frozen: re-anchoring an applied/resolved
+    // thread is meaningless and could resurrect a stale expectedText.
+    if (comment.suggestionAppliedAt || comment.resolvedAt) {
+      throw new BadRequestException(
+        'Cannot re-anchor a suggestion that was already applied or resolved',
+      );
+    }
+    const trimmed = selection.trim();
+    if (trimmed.length === 0) {
+      throw new BadRequestException('The re-anchored selection cannot be empty');
+    }
+    // Same no-op guard as create(): the suggestion must differ from the text it
+    // replaces, or apply becomes indistinguishable from already-applied.
+    if (trimmed === comment.suggestedText.trim()) {
+      throw new BadRequestException(
+        'A suggested edit must differ from the selected text',
+      );
+    }
+
+    // Idempotent: nothing to persist when the anchor already matches.
+    if (comment.selection === selection) {
+      return comment;
+    }
+
+    await this.commentRepo.updateComment({ selection }, comment.id);
+
+    const updatedComment = await this.commentRepo.findById(comment.id, {
+      includeCreator: true,
+      includeResolvedBy: true,
+    });
+
+    // Re-anchoring only corrects stored metadata; it does not change the page
+    // text or the comment body, so no ws broadcast / notification is warranted.
+    return updatedComment;
+  }
+
   /**
    * Apply the suggested edit carried by a top-level inline comment: atomically
    * replace the text under the comment mark in the collaborative document with
@@ -524,7 +594,7 @@ export class CommentService {
         resourceType: AuditResource.COMMENT,
         resourceId: comment.id,
         spaceId: comment.spaceId,
-        metadata: { pageId: comment.pageId },
+        metadata: this.suggestionAuditMetadata(comment, user),
       });
       return { ...updatedComment, outcome: 'resolved' };
     }
@@ -538,7 +608,7 @@ export class CommentService {
       resourceType: AuditResource.COMMENT,
       resourceId: comment.id,
       spaceId: comment.spaceId,
-      metadata: { pageId: comment.pageId },
+      metadata: this.suggestionAuditMetadata(comment, user),
     });
     return settled;
   }
@@ -577,8 +647,10 @@ export class CommentService {
 
       // Auto-resolve the thread. resolveComment handles the resolve mark, its ws
       // broadcast and the resolve notification. Stay defensive on re-entry.
+      let didResolveBroadcast = false;
       if (!comment.resolvedAt) {
         await this.resolveComment(comment, true, user, provenance);
+        didResolveBroadcast = true;
       }
 
       const updatedComment = await this.commentRepo.findById(comment.id, {
@@ -586,18 +658,27 @@ export class CommentService {
         includeResolvedBy: true,
       });
 
-      this.wsService.emitCommentEvent(comment.spaceId, comment.pageId, {
-        operation: 'commentUpdated',
-        pageId: comment.pageId,
-        comment: updatedComment,
-      });
+      // #496 dedup: resolveComment already broadcast `commentResolved` carrying
+      // the fully-enriched row (the applied stamps were persisted above, before
+      // that call, so its re-read reflects them). Emitting `commentUpdated` here
+      // too made the client receive TWO events for one apply. Broadcast the
+      // update ONLY when we did NOT resolve — i.e. the rare re-entry on an
+      // already-resolved thread, where the applied-stamp change still needs a
+      // broadcast and resolveComment did not run.
+      if (!didResolveBroadcast) {
+        this.wsService.emitCommentEvent(comment.spaceId, comment.pageId, {
+          operation: 'commentUpdated',
+          pageId: comment.pageId,
+          comment: updatedComment,
+        });
+      }
 
       this.auditService.log({
         event: AuditEvent.COMMENT_SUGGESTION_APPLIED,
         resourceType: AuditResource.COMMENT,
         resourceId: comment.id,
         spaceId: comment.spaceId,
-        metadata: { pageId: comment.pageId },
+        metadata: this.suggestionAuditMetadata(comment, user),
       });
 
       return { ...updatedComment, outcome: 'resolved' };
@@ -616,7 +697,7 @@ export class CommentService {
       resourceType: AuditResource.COMMENT,
       resourceId: comment.id,
       spaceId: comment.spaceId,
-      metadata: { pageId: comment.pageId },
+      metadata: this.suggestionAuditMetadata(comment, user),
     });
 
     return settled;
@@ -627,14 +708,17 @@ export class CommentService {
    * inline `comment` anchor mark, then ATOMICALLY hard-delete the row only if it
    * is still childless. Shared by the apply/dismiss no-replies branches (#329).
    *
-   * ORDER MATTERS: the anchor mark is removed FIRST and FATALLY (mirrors
-   * applySuggestion, which mutates the doc before writing the DB). The row
-   * delete is irreversible, so if the mark removal fails — including the
-   * COLLAB_DISABLE_REDIS "no live instance" hard-error — we must NOT delete the
-   * row and report success, or the document is left with a permanent orphan
-   * anchor pointing at a comment that no longer exists (the exact data-integrity
-   * bug #329 targets). Let the exception propagate (→ 5xx); the operation is
-   * then repeatable with row + mark still consistent.
+   * ORDER MATTERS (updated #399 → #496): what runs FIRST and FATALLY here is the
+   * mark-removal ENQUEUE (a fast, durable Redis add), NOT the mark op itself.
+   * deleteCommentMark awaits only the enqueue, so a failed add throws BEFORE the
+   * irreversible row delete — the row + mark stay consistent and the operation is
+   * repeatable. The actual anchor strip then runs off the HTTP path in the worker
+   * (idempotent, 3 retries). Only an EXHAUSTED-retries job could leave the doc
+   * with an orphan anchor pointing at a hard-deleted comment (the data-integrity
+   * bug #329 targets); that residual divergence is now self-healed by the
+   * resolve/unresolve mark worker, which strips an orphan mark whenever its
+   * comment row is gone (#496), and it is meanwhile VISIBLE via BullMQ failed-job
+   * metrics rather than a silently-swallowed warn.
    *
    * RACE (#338 F4): the caller read `hasChildren` BEFORE the (slow) mark
    * removal, so a reply can land in that window. `comments.parent_comment_id` is
@@ -732,6 +816,27 @@ export class CommentService {
     return this.generalQueue.add(QueueJob.COMMENT_MARK_UPDATE, jobData);
   }
 
+  /**
+   * Build the audit metadata for a suggestion apply/dismiss decision (#496).
+   * The subject comment is HARD-DELETED on the childless path, so the audit row
+   * is the only surviving record — capture the decision's substance (what was
+   * suggested, the anchored text it replaced, who authored it, who decided)
+   * before the row can vanish. `decidedBy` is the acting user; `commentAuthor`
+   * is the suggestion's creator.
+   */
+  private suggestionAuditMetadata(
+    comment: Comment,
+    user: User,
+  ): Record<string, any> {
+    return {
+      pageId: comment.pageId,
+      suggestedText: comment.suggestedText ?? null,
+      selection: comment.selection ?? null,
+      commentAuthor: comment.creatorId ?? null,
+      decidedBy: user.id,
+    };
+  }
+
   private async queueCommentNotification(
     content: any,
     oldMentionIds: string[],

apps/server/src/core/comment/dto/resync-suggestion-anchor.dto.ts

@@ -0,0 +1,20 @@
+import { IsString, IsUUID, MaxLength, MinLength } from 'class-validator';
+
+/**
+ * #496: after the MCP client anchors a suggestion in the LIVE collab doc, it
+ * re-reads the exact substring under the new mark and syncs it here as the
+ * comment's stored `selection` (== apply-time expectedText). Fixes the perpetual
+ * 409 where expectedText came from a debounced REST snapshot while the mark sat
+ * in the live doc.
+ */
+export class ResyncSuggestionAnchorDto {
+  @IsUUID()
+  commentId: string;
+
+  // The raw substring the mark now covers in the live document. Bounded like the
+  // create-time selection (2000) so a legitimate anchored span is never cut.
+  @IsString()
+  @MinLength(1)
+  @MaxLength(2000)
+  selection: string;
+}

apps/server/src/integrations/audit/audit.module.ts

@@ -1,14 +1,19 @@
 import { Global, Module } from '@nestjs/common';
-import { AUDIT_SERVICE, NoopAuditService } from './audit.service';
+import { AUDIT_SERVICE } from './audit.service';
+import { DatabaseAuditService } from './database-audit.service';
 
+// #496: bind the audit token to a real DB-backed trail (was NoopAuditService,
+// which silently dropped every event). Kysely (@Global DatabaseModule) and
+// ClsService (@Global ClsModule) are both globally available, so this module
+// needs no extra imports.
 @Global()
 @Module({
   providers: [
     {
       provide: AUDIT_SERVICE,
-      useClass: NoopAuditService,
+      useClass: DatabaseAuditService,
     },
   ],
   exports: [AUDIT_SERVICE],
 })
-export class NoopAuditModule {}
+export class AuditModule {}

apps/server/src/integrations/audit/database-audit.service.spec.ts

@@ -0,0 +1,198 @@
+import { Logger } from '@nestjs/common';
+import { DatabaseAuditService } from './database-audit.service';
+import { AUDIT_CONTEXT_KEY } from '../../common/middlewares/audit-context.middleware';
+import { AuditEvent, AuditResource } from '../../common/events/audit-events';
+
+/**
+ * Observable-property coverage for the DB-backed audit trail (#496): every
+ * assertion pins what actually reaches the `audit` table (or that nothing does),
+ * driven through a chainable Kysely mock that captures the inserted rows.
+ */
+describe('DatabaseAuditService', () => {
+  function makeService(clsContext: any) {
+    const inserted: any[] = [];
+    const updated: any[] = [];
+    let failNextInsert = false;
+
+    const db: any = {
+      insertInto: jest.fn(() => ({
+        values: jest.fn((rows: any) => ({
+          execute: jest.fn(async () => {
+            if (failNextInsert) {
+              failNextInsert = false;
+              throw new Error('boom');
+            }
+            inserted.push(rows);
+          }),
+        })),
+      })),
+      updateTable: jest.fn(() => ({
+        set: jest.fn((patch: any) => ({
+          where: jest.fn(() => ({
+            execute: jest.fn(async () => {
+              updated.push(patch);
+            }),
+          })),
+        })),
+      })),
+    };
+
+    const store: Record<string, any> = { [AUDIT_CONTEXT_KEY]: clsContext };
+    const cls: any = {
+      get: jest.fn((key: string) => store[key]),
+      set: jest.fn((key: string, val: any) => {
+        store[key] = val;
+      }),
+    };
+
+    const service = new DatabaseAuditService(db, cls);
+    return {
+      service,
+      inserted,
+      updated,
+      cls,
+      store,
+      failInsert: () => {
+        failNextInsert = true;
+      },
+    };
+  }
+
+  const applyPayload = () => ({
+    event: AuditEvent.COMMENT_SUGGESTION_APPLIED,
+    resourceType: AuditResource.COMMENT,
+    resourceId: 'c-1',
+    spaceId: 'space-1',
+    metadata: { pageId: 'p-1', suggestedText: 'new', decidedBy: 'u-2' },
+  });
+
+  const ctx = (over?: any) => ({
+    workspaceId: 'ws-1',
+    actorId: 'u-2',
+    actorType: 'user',
+    ipAddress: '10.0.0.1',
+    userAgent: 'jest',
+    ...over,
+  });
+
+  it('log() persists a row with the CLS context merged onto the payload', async () => {
+    const { service, inserted } = makeService(ctx());
+    service.log(applyPayload());
+    // log() is fire-and-forget; flush the microtask queue.
+    await Promise.resolve();
+    await Promise.resolve();
+
+    expect(inserted).toHaveLength(1);
+    expect(inserted[0]).toMatchObject({
+      workspaceId: 'ws-1',
+      actorId: 'u-2',
+      actorType: 'user',
+      event: AuditEvent.COMMENT_SUGGESTION_APPLIED,
+      resourceType: AuditResource.COMMENT,
+      resourceId: 'c-1',
+      spaceId: 'space-1',
+      ipAddress: '10.0.0.1',
+      metadata: { pageId: 'p-1', suggestedText: 'new', decidedBy: 'u-2' },
+    });
+  });
+
+  it('log() is a no-op when there is no workspace in scope', async () => {
+    const { service, inserted } = makeService(ctx({ workspaceId: null }));
+    service.log(applyPayload());
+    await Promise.resolve();
+    expect(inserted).toHaveLength(0);
+  });
+
+  it('log() drops EXCLUDED_AUDIT_EVENTS (e.g. comment.created)', async () => {
+    const { service, inserted } = makeService(ctx());
+    service.log({
+      event: AuditEvent.COMMENT_CREATED,
+      resourceType: AuditResource.COMMENT,
+      resourceId: 'c-1',
+      spaceId: 'space-1',
+    });
+    await Promise.resolve();
+    await Promise.resolve();
+    expect(inserted).toHaveLength(0);
+  });
+
+  it('a failed insert is swallowed with a warn and floats no rejection (audit is a side-record)', async () => {
+    // This pins the load-bearing swallow inside persist(). Because log() is
+    // fire-and-forget (`void this.persist(...)`), it always returns synchronously
+    // without throwing — so `not.toThrow()` alone would stay green even if the
+    // try/catch were removed. We instead observe the two effects the catch is
+    // responsible for: a warn IS emitted, and NO unhandled rejection floats.
+    // Removing persist()'s try/catch reddens both assertions (warn count 0 + a
+    // captured rejection).
+    const warnSpy = jest
+      .spyOn(Logger.prototype, 'warn')
+      .mockImplementation(() => undefined as any);
+    const rejections: unknown[] = [];
+    const onRejection = (err: unknown) => rejections.push(err);
+    process.on('unhandledRejection', onRejection);
+    try {
+      const { service, failInsert } = makeService(ctx());
+      failInsert();
+      expect(() => service.log(applyPayload())).not.toThrow();
+      // Flush microtasks so the rejected insert settles, then give any floated
+      // rejection a macrotask tick to be reported by the runtime.
+      await Promise.resolve();
+      await Promise.resolve();
+      await new Promise((r) => setImmediate(r));
+
+      expect(warnSpy).toHaveBeenCalledTimes(1);
+      expect(String(warnSpy.mock.calls[0][0])).toContain(
+        'Failed to persist audit event',
+      );
+      expect(rejections).toHaveLength(0);
+    } finally {
+      process.off('unhandledRejection', onRejection);
+      warnSpy.mockRestore();
+    }
+  });
+
+  it('logWithContext() persists with an explicit (non-request) context', async () => {
+    const { service, inserted } = makeService(undefined);
+    service.logWithContext(applyPayload(), ctx({ actorType: 'system' }) as any);
+    await Promise.resolve();
+    await Promise.resolve();
+    expect(inserted).toHaveLength(1);
+    expect(inserted[0].actorType).toBe('system');
+    expect(inserted[0].workspaceId).toBe('ws-1');
+  });
+
+  it('logBatchWithContext() inserts only non-excluded events', async () => {
+    const { service, inserted } = makeService(undefined);
+    service.logBatchWithContext(
+      [
+        applyPayload(),
+        {
+          event: AuditEvent.COMMENT_CREATED,
+          resourceType: AuditResource.COMMENT,
+          resourceId: 'c-2',
+        },
+      ],
+      ctx() as any,
+    );
+    await Promise.resolve();
+    await Promise.resolve();
+    // Batch is a single insert call carrying only the applied event.
+    expect(inserted).toHaveLength(1);
+    expect(inserted[0]).toHaveLength(1);
+    expect(inserted[0][0].event).toBe(AuditEvent.COMMENT_SUGGESTION_APPLIED);
+  });
+
+  it('setActorId / setActorType mutate the ambient CLS context', () => {
+    const { service, store } = makeService(ctx({ actorId: null }));
+    service.setActorId('u-9');
+    service.setActorType('api_key');
+    expect(store[AUDIT_CONTEXT_KEY].actorId).toBe('u-9');
+    expect(store[AUDIT_CONTEXT_KEY].actorType).toBe('api_key');
+  });
+
+  it('updateRetention() writes the workspace retention window', async () => {
+    const { service, updated } = makeService(ctx());
+    await service.updateRetention('ws-1', 30);
+    expect(updated).toEqual([{ auditRetentionDays: 30 }]);
+  });
+});

apps/server/src/integrations/audit/database-audit.service.ts

@@ -0,0 +1,160 @@
+import { Injectable, Logger } from '@nestjs/common';
+import { InjectKysely } from 'nestjs-kysely';
+import { ClsService } from 'nestjs-cls';
+import { KyselyDB } from '@docmost/db/types/kysely.types';
+import {
+  AuditContext,
+  AUDIT_CONTEXT_KEY,
+} from '../../common/middlewares/audit-context.middleware';
+import {
+  AuditLogPayload,
+  ActorType,
+  EXCLUDED_AUDIT_EVENTS,
+} from '../../common/events/audit-events';
+import { AuditLogContext, IAuditService } from './audit.service';
+
+/**
+ * Minimal DB-backed audit trail (#496). Replaces NoopAuditService so that
+ * decision-bearing events — notably comment.suggestion_applied /
+ * comment.suggestion_dismissed, whose subject comment is HARD-DELETED on the
+ * childless path — leave a durable record of who decided what. Without this the
+ * events were emitted (comment.service / *.controller) but swallowed, so an
+ * applied/dismissed suggestion was unrecoverable once the row was gone.
+ *
+ * Rows land in the pre-existing `audit` table (migration 20260228T223532). The
+ * per-request actor/workspace/ip come from the CLS AuditContext populated by
+ * AuditContextMiddleware + AuditActorInterceptor; callers that run OUTSIDE a
+ * request (queue workers, imports) pass an explicit context via
+ * logWithContext / logBatchWithContext.
+ *
+ * Audit is a side-record: a write failure MUST NOT break the originating
+ * request, so every persistence path swallows its error with a warn. Events in
+ * EXCLUDED_AUDIT_EVENTS (high-volume/low-signal) are dropped.
+ */
+@Injectable()
+export class DatabaseAuditService implements IAuditService {
+  private readonly logger = new Logger(DatabaseAuditService.name);
+
+  constructor(
+    @InjectKysely() private readonly db: KyselyDB,
+    private readonly cls: ClsService,
+  ) {}
+
+  /**
+   * Persist a single event using the ambient request-scoped AuditContext. A
+   * no-op when there is no workspace in scope (the table's workspace_id is NOT
+   * NULL) or the event is excluded. Fire-and-forget: the returned promise is not
+   * awaited by hot callers, and its rejection is swallowed here.
+   */
+  log(payload: AuditLogPayload): void {
+    const context = this.cls?.get<AuditContext>(AUDIT_CONTEXT_KEY);
+    if (!context?.workspaceId) {
+      // No workspace in scope — nothing we can attribute the row to. This is
+      // expected for events emitted outside an HTTP request; those callers must
+      // use logWithContext instead.
+      return;
+    }
+    void this.persist(payload, {
+      workspaceId: context.workspaceId,
+      actorId: context.actorId ?? undefined,
+      actorType: context.actorType,
+      ipAddress: context.ipAddress ?? undefined,
+      userAgent: context.userAgent ?? undefined,
+    });
+  }
+
+  /** Persist a single event with an explicit (non-request) context. */
+  logWithContext(payload: AuditLogPayload, context: AuditLogContext): void {
+    if (!context?.workspaceId) return;
+    void this.persist(payload, context);
+  }
+
+  /** Persist a batch of events sharing one explicit context (imports). */
+  logBatchWithContext(
+    payloads: AuditLogPayload[],
+    context: AuditLogContext,
+  ): void {
+    if (!context?.workspaceId || payloads.length === 0) return;
+    const rows = payloads
+      .filter((p) => !EXCLUDED_AUDIT_EVENTS.has(p.event))
+      .map((p) => this.toRow(p, context));
+    if (rows.length === 0) return;
+    this.db
+      .insertInto('audit')
+      .values(rows)
+      .execute()
+      .catch((err: any) =>
+        this.logger.warn(`Failed to persist ${rows.length} audit events: ${err?.message}`),
+      );
+  }
+
+  /** Update the ambient request actor (e.g. after login resolves the user). */
+  setActorId(actorId: string): void {
+    const context = this.cls?.get<AuditContext>(AUDIT_CONTEXT_KEY);
+    if (context) {
+      context.actorId = actorId;
+      this.cls.set(AUDIT_CONTEXT_KEY, context);
+    }
+  }
+
+  /** Update the ambient request actor type (user | system | api_key). */
+  setActorType(actorType: ActorType): void {
+    const context = this.cls?.get<AuditContext>(AUDIT_CONTEXT_KEY);
+    if (context) {
+      context.actorType = actorType;
+      this.cls.set(AUDIT_CONTEXT_KEY, context);
+    }
+  }
+
+  /** Persist a workspace's audit-log retention window (days). */
+  async updateRetention(
+    workspaceId: string,
+    retentionDays: number,
+  ): Promise<void> {
+    try {
+      await this.db
+        .updateTable('workspaces')
+        .set({ auditRetentionDays: retentionDays })
+        .where('id', '=', workspaceId)
+        .execute();
+    } catch (err: any) {
+      this.logger.warn(
+        `Failed to update audit retention for workspace ${workspaceId}: ${err?.message}`,
+      );
+    }
+  }
+
+  private async persist(
+    payload: AuditLogPayload,
+    context: AuditLogContext,
+  ): Promise<void> {
+    if (EXCLUDED_AUDIT_EVENTS.has(payload.event)) return;
+    try {
+      await this.db
+        .insertInto('audit')
+        .values(this.toRow(payload, context))
+        .execute();
+    } catch (err: any) {
+      // Audit is a side-record; never let a failed write surface to the caller.
+      this.logger.warn(
+        `Failed to persist audit event ${payload.event}: ${err?.message}`,
+      );
+    }
+  }
+
+  private toRow(payload: AuditLogPayload, context: AuditLogContext) {
+    return {
+      workspaceId: context.workspaceId,
+      actorId: context.actorId ?? null,
+      actorType: context.actorType ?? 'user',
+      event: payload.event,
+      resourceType: payload.resourceType,
+      resourceId: payload.resourceId ?? null,
+      spaceId: payload.spaceId ?? null,
+      // jsonb columns: node-postgres serializes plain objects to JSON.
+      changes: payload.changes ? (payload.changes as any) : null,
+      metadata: payload.metadata ? (payload.metadata as any) : null,
+      ipAddress: context.ipAddress ?? null,
+    };
+  }
+}

apps/server/src/integrations/export/internal-link-parity.spec.ts

@@ -0,0 +1,70 @@
+import { INTERNAL_LINK_REGEX } from './utils';
+import { isInternalPagePath } from '@docmost/prosemirror-markdown';
+
+/**
+ * Cross-package DRIFT GUARD for the internal-link subset invariant (#522).
+ *
+ * The client-side `isInternalPagePath`
+ * (`packages/prosemirror-markdown/src/lib/internal-links.ts`) promotes a markdown
+ * link to `internal: true` on import. Every link it marks internal MUST be one
+ * the server would backlink and export-rewrite — i.e. the client matcher MUST be
+ * a STRICT SUBSET of the server's canonical `INTERNAL_LINK_REGEX`
+ * (`./utils.ts`). If the client ever accepts a path the server rejects, that link
+ * is stored internal but silently dropped from the backlink graph and broken on
+ * export — the exact bug #522 fixed.
+ *
+ * This spec is the load-bearing guard: it imports the LIVE server regex AND the
+ * LIVE `isInternalPagePath` (no hand-copied regex on either side). A narrowing of
+ * EITHER — most dangerously the server regex — reddens here. The in-package
+ * accept/reject test documents the client's behaviour but cannot see the server
+ * regex; this top-layer spec is what makes the subset relation mechanical
+ * (AGENTS.md rule #7: a CI test that fails on drift of the source of truth).
+ */
+describe('internal-link subset parity (client isInternalPagePath ⊆ server INTERNAL_LINK_REGEX)', () => {
+  // Every path the CLIENT accepts. Kept deliberately broad across the risky
+  // dimensions — the slug charset (digits, hyphens, mixed case, leading/trailing
+  // hyphen), the space charset, and the optional trailing slash — so a narrowing
+  // of the server regex on any of them reddens the subset assertion below.
+  const CLIENT_ACCEPTS = [
+    '/s/eng/p/abc123',
+    '/s/eng/p/abc123/', // trailing slash
+    '/s/My-Space/p/A-b-C-9', // hyphens + mixed case in both segments
+    '/s/x/p/z', // shortest
+    '/s/eng/p/my-page-abc123', // slug with hyphens (extractPageSlugId shape)
+    '/s/eng/p/-lead', // leading hyphen in slug
+    '/s/eng/p/trail-', // trailing hyphen in slug
+    '/s/eng/p/0123456789', // all-digit slug
+    '/s/a.b/p/abc', // dot in the SPACE segment (space charset is [^/]+)
+    '/s/space with space/p/abc', // space char in the SPACE segment
+  ];
+
+  it('every corpus path is actually client-accepted (guards the corpus itself)', () => {
+    // If a path here stopped being client-accepted the subset test would pass
+    // vacuously; assert acceptance up front so the corpus stays meaningful. The
+    // filter-to-empty form names the offending paths on failure.
+    const notAccepted = CLIENT_ACCEPTS.filter((h) => !isInternalPagePath(h));
+    expect(notAccepted).toEqual([]);
+  });
+
+  it('every client-accepted path also matches the LIVE server regex (subset)', () => {
+    // The mechanical drift guard: narrow the server INTERNAL_LINK_REGEX and at
+    // least one hyphen/charset/structure case appears here.
+    const notInServer = CLIENT_ACCEPTS.filter((h) => !INTERNAL_LINK_REGEX.test(h));
+    expect(notInServer).toEqual([]);
+  });
+
+  it('the client rejects forbidden-slug-char paths the server also rejects', () => {
+    // Documents the subset BOUNDARY: correct shape, forbidden slug char. The
+    // client and the LIVE server regex must agree on rejection.
+    const forbidden = [
+      '/s/eng/p/abc.def',
+      '/s/eng/p/abc_def',
+      '/s/eng/p/abc%20',
+      '/s/eng/p/abc~x',
+    ];
+    const clientAccepts = forbidden.filter((h) => isInternalPagePath(h));
+    const serverAccepts = forbidden.filter((h) => INTERNAL_LINK_REGEX.test(h));
+    expect(clientAccepts).toEqual([]);
+    expect(serverAccepts).toEqual([]);
+  });
+});

apps/server/src/integrations/queue/processors/general-queue.processor.comment-mark.spec.ts

@@ -139,13 +139,19 @@ describe('GeneralQueueProcessor — COMMENT_MARK_UPDATE (#399)', () => {
     );
   });
 
-  it('skips (no throw) when the comment row has vanished', async () => {
+  it('reconcile (#496): comment row vanished → strips the orphan anchor mark', async () => {
     const { proc, collaborationGateway, commentRepo } = makeProc();
     commentRepo.findById.mockResolvedValue(undefined);
 
     await expect(
       proc.process(job({ ...base, action: 'resolve', ts: 1000 })),
     ).resolves.toBeUndefined();
-    expect(collaborationGateway.handleYjsEvent).not.toHaveBeenCalled();
+    // A resolve/unresolve mark job whose comment row is gone leaves a silent
+    // orphan; the worker self-heals by stripping the anchor instead of returning.
+    expect(collaborationGateway.handleYjsEvent).toHaveBeenCalledWith(
+      'deleteCommentMark',
+      'page.page-1',
+      { commentId: 'c-1', user: { id: 'user-1' } },
+    );
   });
 });

apps/server/src/integrations/queue/processors/general-queue.processor.ts

@@ -95,7 +95,8 @@ export class GeneralQueueProcessor
    * #399: apply a comment's inline-mark mirror in the collab Y.Doc, off the HTTP
    * critical path. Runs the SAME gateway path the synchronous comment.service
    * code used (byte-identical mark op):
-   *   - resolve / unresolve → resolveCommentMark (flip the `resolved` attribute);
+   *   - resolve / unresolve → resolveCommentMark (flip the `resolved` attribute),
+   *     OR strip an orphan anchor when the comment row has vanished (#496);
    *   - delete → deleteCommentMark (strip the ephemeral-suggestion anchor #329).
    * The op is idempotent, so a BullMQ retry is safe. Throwing propagates to
    * WorkerHost → the job is retried and, on exhaustion, surfaces in failed-job
@@ -133,7 +134,18 @@ export class GeneralQueueProcessor
     // of this resolve), skip it rather than flip the mark to a stale state.
     const comment = await this.commentRepo.findById(commentId);
     if (!comment) {
-      // The comment vanished (e.g. hard-deleted) → nothing left to mirror.
+      // #496 reconcile: the comment row is GONE (e.g. an ephemeral apply/dismiss
+      // hard-deleted it while this resolve/unresolve mark job sat in the queue),
+      // but its inline anchor may still live in the doc — a silent orphan mark
+      // pointing at a comment that no longer exists. Self-heal by stripping it
+      // instead of just returning: this closes the divergence the fire-and-forget
+      // resolve/unresolve enqueue (comment.service resolveComment) could leave.
+      // Idempotent — deleteCommentMark on an already-absent mark is a no-op.
+      await this.getCollaborationGateway().handleYjsEvent(
+        'deleteCommentMark',
+        documentName,
+        { commentId, user },
+      );
       return;
     }
     const wantResolved = action === 'resolve';

packages/mcp/src/client/comments.ts

@@ -450,6 +450,12 @@ export function CommentsMixin<TBase extends GConstructor<DocmostClientContext>>(
     // can surface the closest-block / spans-multiple-blocks hint built from the
     // LIVE document (the pre-check page is not in scope there).
     let liveNotFoundError: Error | null = null;
+    // #496: the RAW substring the mark actually covers in the LIVE doc. The
+    // stored selection (payload.selection) came from a DEBOUNCED REST snapshot,
+    // which can differ from the live doc — and apply compares the marked live
+    // text to the stored selection strictly, so a stale snapshot 409s on EVERY
+    // apply. Captured here (same doc version the mark is set in) and synced below.
+    let liveAnchoredSelection: string | null = null;
     try {
       const collabToken = await this.getCollabTokenWithReauth();
       // Open the collab doc by the canonical UUID, never the slugId (#260). The
@@ -489,6 +495,12 @@ export function CommentsMixin<TBase extends GConstructor<DocmostClientContext>>(
           }
           if (applyAnchorInDoc(doc, selection as string, newCommentId)) {
             anchored = true;
+            // For a suggestion, re-read the exact substring now under the mark
+            // (the mark is an attribute, so it does not change the raw text) to
+            // sync as the stored expectedText after the mutation resolves.
+            if (hasSuggestion) {
+              liveAnchoredSelection = getAnchoredText(doc, selection as string);
+            }
             return doc;
           }
           // Selection text not found in the LIVE document: abort the write. The
@@ -527,6 +539,36 @@ export function CommentsMixin<TBase extends GConstructor<DocmostClientContext>>(
       );
     }
 
+    // #496: sync the stored selection (== apply-time expectedText) to the RAW
+    // substring the mark actually covers in the LIVE doc when it diverged from
+    // the debounced REST snapshot we stored at create time. Without this, apply
+    // strictly compares the marked live text to a stale stored selection and
+    // 409s every time. Best-effort: the comment is already correctly anchored, so
+    // a resync failure must NOT roll it back — it only risks a later apply 409,
+    // which we surface as a soft warning.
+    if (
+      hasSuggestion &&
+      liveAnchoredSelection != null &&
+      liveAnchoredSelection !== payload.selection
+    ) {
+      try {
+        await this.client.post("/comments/resync-suggestion-anchor", {
+          commentId: newCommentId,
+          selection: liveAnchoredSelection,
+        });
+        // Reflect the corrected anchor in the returned comment.
+        if (result.data) result.data.selection = liveAnchoredSelection;
+      } catch (e) {
+        if (process.env.DEBUG) {
+          console.error("Failed to resync suggestion anchor:", e);
+        }
+        result.warning =
+          "The suggestion was anchored, but its stored selection could not be " +
+          "synced to the live document; applying it may report a conflict if the " +
+          "text changed. Re-create the suggestion if Apply fails.";
+      }
+    }
+
     // Soft warning (like editPageText): the selection only matched after
     // stripping markdown, so the caller likely quoted a styled fragment.
     if (anchorNormalized) {

packages/mcp/test/mock/create-comment.test.mjs

@@ -553,6 +553,189 @@ test("suggestedText: the stored selection is the doc's RAW typographic substring
   assert.equal(createPayload.suggestedText, "goodbye");
 });
 
+// -----------------------------------------------------------------------------
+// 8b) #496: the DEBOUNCED REST snapshot (pages/info) DIFFERS from the LIVE collab
+//     doc (mutatePage) — the doc moved on in the debounce window. The stored
+//     selection is captured from the snapshot at create time, but the mark is set
+//     in the live doc, so apply would 409 forever. After anchoring, the client
+//     re-reads the RAW substring under the mark from the LIVE doc and POSTs it to
+//     /comments/resync-suggestion-anchor so the stored expectedText matches.
+// -----------------------------------------------------------------------------
+test("suggestion: re-syncs the stored selection to the LIVE doc substring when the REST snapshot lagged", async () => {
+  let createPayload = null;
+  let resyncPayload = null;
+
+  const { baseURL } = await spawn(async (req, res) => {
+    const raw = await readBody(req);
+    if (req.url === "/api/auth/login") {
+      sendJson(res, 200, { success: true }, {
+        "Set-Cookie": "authToken=t; Path=/; HttpOnly",
+      });
+      return;
+    }
+    if (req.url === "/api/pages/info") {
+      // DEBOUNCED snapshot: ASCII quotes (stale — the live doc has since been
+      // typographically corrected).
+      sendJson(res, 200, {
+        data: {
+          id: "33333333-3333-3333-3333-333333333333",
+          content: {
+            type: "doc",
+            content: [
+              {
+                type: "paragraph",
+                content: [{ type: "text", text: 'he said "hello" loudly' }],
+              },
+            ],
+          },
+        },
+      });
+      return;
+    }
+    if (req.url === "/api/comments/create") {
+      createPayload = JSON.parse(raw);
+      sendJson(res, 200, {
+        data: {
+          id: "cmt-resync-1",
+          content: createPayload.content,
+          selection: createPayload.selection,
+          suggestedText: createPayload.suggestedText,
+          type: createPayload.type,
+        },
+      });
+      return;
+    }
+    if (req.url === "/api/comments/resync-suggestion-anchor") {
+      resyncPayload = JSON.parse(raw);
+      sendJson(res, 200, { data: { id: "cmt-resync-1" } });
+      return;
+    }
+    sendJson(res, 404, { message: "not found" });
+  });
+
+  class TestClient extends DocmostClient {
+    async getCollabTokenWithReauth() {
+      return "collab-token";
+    }
+    async resolvePageId() {
+      return "33333333-3333-3333-3333-333333333333";
+    }
+    async mutatePage(pageId, collabToken, apiUrl, transform) {
+      // LIVE doc: SMART quotes (what the mark is actually set over).
+      const doc = {
+        type: "doc",
+        content: [
+          {
+            type: "paragraph",
+            content: [{ type: "text", text: "he said “hello” loudly" }],
+          },
+        ],
+      };
+      const out = transform(doc);
+      return { doc: out, verify: { ok: true } };
+    }
+  }
+
+  const client = new TestClient(baseURL, "user@example.com", "pw");
+
+  const result = await client.createComment(
+    "33333333-3333-3333-3333-333333333333",
+    "please change",
+    "inline",
+    '"hello"', // ASCII quotes
+    undefined,
+    "goodbye",
+  );
+
+  assert.equal(result.success, true);
+  assert.equal(result.anchored, true);
+  // Create stored the STALE snapshot substring (ASCII quotes).
+  assert.equal(createPayload.selection, '"hello"');
+  // …then the client re-synced to the LIVE marked substring (smart quotes).
+  assert.ok(resyncPayload, "/comments/resync-suggestion-anchor must be called");
+  assert.equal(resyncPayload.commentId, "cmt-resync-1");
+  assert.equal(resyncPayload.selection, "“hello”");
+  // The returned comment reflects the corrected anchor.
+  assert.equal(result.data.selection, "“hello”");
+});
+
+// -----------------------------------------------------------------------------
+// 8c) #496: when the REST snapshot and the LIVE doc AGREE, no resync round-trip
+//     is made (the common case must stay a single write).
+// -----------------------------------------------------------------------------
+test("suggestion: no resync call when the snapshot already matches the live doc", async () => {
+  let resyncCalls = 0;
+
+  const { baseURL } = await spawn(async (req, res) => {
+    const raw = await readBody(req);
+    if (req.url === "/api/auth/login") {
+      sendJson(res, 200, { success: true }, {
+        "Set-Cookie": "authToken=t; Path=/; HttpOnly",
+      });
+      return;
+    }
+    if (req.url === "/api/pages/info") {
+      sendJson(res, 200, {
+        data: {
+          id: "44444444-4444-4444-4444-444444444444",
+          content: {
+            type: "doc",
+            content: [
+              { type: "paragraph", content: [{ type: "text", text: "Hello brave world" }] },
+            ],
+          },
+        },
+      });
+      return;
+    }
+    if (req.url === "/api/comments/create") {
+      const p = JSON.parse(raw);
+      sendJson(res, 200, {
+        data: { id: "cmt-nosync-1", content: p.content, selection: p.selection, suggestedText: p.suggestedText, type: p.type },
+      });
+      return;
+    }
+    if (req.url === "/api/comments/resync-suggestion-anchor") {
+      resyncCalls++;
+      sendJson(res, 200, { data: {} });
+      return;
+    }
+    sendJson(res, 404, { message: "not found" });
+  });
+
+  class TestClient extends DocmostClient {
+    async getCollabTokenWithReauth() {
+      return "collab-token";
+    }
+    async resolvePageId() {
+      return "44444444-4444-4444-4444-444444444444";
+    }
+    async mutatePage(pageId, collabToken, apiUrl, transform) {
+      const doc = {
+        type: "doc",
+        content: [
+          { type: "paragraph", content: [{ type: "text", text: "Hello brave world" }] },
+        ],
+      };
+      const out = transform(doc);
+      return { doc: out, verify: { ok: true } };
+    }
+  }
+
+  const client = new TestClient(baseURL, "user@example.com", "pw");
+  const result = await client.createComment(
+    "44444444-4444-4444-4444-444444444444",
+    "rename",
+    "inline",
+    "brave",
+    undefined,
+    "bold",
+  );
+
+  assert.equal(result.anchored, true);
+  assert.equal(resyncCalls, 0, "matching snapshot must NOT trigger a resync round-trip");
+});
+
 // -----------------------------------------------------------------------------
 // 8) #408: a not-found selection error QUOTES the closest block text so the
 //    model can self-correct instead of blind-retrying.

packages/prosemirror-markdown/src/lib/canonicalize.ts

@@ -25,13 +25,22 @@
  * drop any attr whose value equals its known schema default. A non-default
  * value (e.g. `orderedList.start: 5`) is NOT a default, so it is KEPT.
  *
- * Every entry below was read from `packages/docmost-client/src/lib/
+ * Every entry below — with the single documented exception of the `link.internal`
+ * marker (see its own bullet) — was read from `packages/docmost-client/src/lib/
  * docmost-schema.ts` (the line refs are the exact `default:` declarations) and
  * confirmed to be materialized by an export→import→export round-trip:
  *   - mark `link`    target / rel  — DocmostAttributes + StarterKit link.
  *       StarterKit's link extension defaults `target: "_blank"` and
  *       `rel: "noopener noreferrer nofollow"`; both materialize on import
  *       (empirically confirmed) even when the source had only `href`.
+ *   - mark `link`    internal (#522) — the ONE editor-sourced, NOT-import-
+ *       materialized default here. Its source is `editor-ext/src/lib/link.ts`
+ *       (default `internal: false`), not docmost-schema, and import does NOT
+ *       materialize it (an imported external link leaves `internal` absent/null,
+ *       never `false`). It is listed so that the editor's stored `internal:false`
+ *       normalizes to the same "external" canon as absent/null (`false ≡ absent`),
+ *       keeping a stored external link canonically equal to its re-import. The
+ *       load-bearing `internal:true` is NON-default and therefore KEPT.
  *   - mark `comment` resolved      — docmost-schema.ts L213-214 (`default: false`).
  *   - node `orderedList` start     — provided by StarterKit's orderedList
  *       (`default: 1`); materializes on import (empirically confirmed).
@@ -56,6 +65,14 @@ const KNOWN_DEFAULTS: Record<string, Record<string, unknown>> = {
   link: {
     target: "_blank",
     rel: "noopener noreferrer nofollow",
+    // Editor-authored EXTERNAL links store `internal: false` (editor-ext link
+    // default `packages/editor-ext/src/lib/link.ts`), while an imported external
+    // link leaves `internal` absent/null. Both mean "external", so `internal:
+    // false` must normalize away exactly like `null`/absent — otherwise a stored
+    // `internal:false` link diverges from its re-import under
+    // `docsCanonicallyEqual` (false !== null). The internal marker `internal:true`
+    // is NON-default, so it is KEPT and survives canonicalization (#522 §11).
+    internal: false,
   },
   comment: {
     resolved: false,

packages/prosemirror-markdown/src/lib/index.ts

@@ -23,6 +23,11 @@ export {
   markdownToProseMirrorSync,
 } from "./markdown-to-prosemirror.js";
 
+// Pure primitive: detect the unambiguously-internal wiki-page link path
+// (`/s/<space>/p/<slug>`) and promote such link marks to their native internal
+// form during import (#522). A strict subset of the server's INTERNAL_LINK_REGEX.
+export { isInternalPagePath, markInternalLinks } from "./internal-links.js";
+
 // The Docmost tiptap schema mirror. Exposed so consumers (and the sync
 // engine's schema-validity regression tests) can build the exact ProseMirror
 // schema the converter targets.

packages/prosemirror-markdown/src/lib/internal-links.ts

@@ -0,0 +1,112 @@
+/**
+ * Detect and mark unambiguously-internal wiki-page links during markdown import.
+ *
+ * WHY THIS EXISTS
+ * ---------------
+ * The markdown converter (`markdownToProseMirrorSync`) materializes every link
+ * with StarterKit's external defaults — `internal: null`, `target: "_blank"`,
+ * `rel: "noopener noreferrer nofollow"`. A markdown link that points at an
+ * internal wiki page written in its host-less, root-relative form
+ * (`[text](/s/<space>/p/<slugId>)`) is therefore stored as EXTERNAL: it opens in
+ * a new tab, gets no hover-preview, and is invisible to the backlink graph
+ * (`extractInternalLinkSlugIds` only counts links whose mark carries
+ * `internal: true`). This module supplies the pure primitive that a post-walk in
+ * the converter uses to promote such links to their native internal form.
+ *
+ * WHERE THE CANON LIVES (a strict subset, on purpose)
+ * ---------------------------------------------------
+ * The authoritative definition of "what is an internal link path" is the
+ * server's `INTERNAL_LINK_REGEX`
+ * (`apps/server/src/integrations/export/utils.ts`):
+ *   /^(https?:\/\/)?([^\/]+)?(\/s\/([^\/]+)\/)?p\/([a-zA-Z0-9-]+)\/?$/
+ * This package is a lower layer than the server app and cannot import it, so the
+ * matcher below is a DOCUMENTED, STRICT SUBSET of that regex: it accepts only the
+ * space-qualified, root-relative page path `/s/<space>/p/<slug>` (with optional
+ * trailing slash). Because it is a subset, every link we mark internal here is
+ * guaranteed to also satisfy the server regex, hence guaranteed backlink-able and
+ * export-rewritable. A unit test pins the exact accept/reject set as the guard
+ * against drift.
+ *
+ * FAIL-TOWARD-EXTERNAL
+ * --------------------
+ * We mark a link internal ONLY on an unambiguous anchored match. Any ambiguity
+ * (a scheme/host, `#anchor`, `?query`, a space-less `/p/<slug>`, a relative
+ * `p/<slug>`, a non-string href) leaves the link external. A false-external is a
+ * soft degradation (a new tab); a false-internal would produce broken SPA
+ * navigation, so "external" is the conservative default.
+ *
+ * Precedent for the target internal shape: the file importer already promotes
+ * internal anchors (`apps/server/src/integrations/import/utils/import-formatter.ts`
+ * `$a.attr('data-internal','true')`) and editor-ext reads it
+ * (`packages/editor-ext/src/lib/link.ts`).
+ */
+
+/**
+ * Matches ONLY the space-qualified, root-relative internal page path
+ * `/s/<space>/p/<slug>` (with optional trailing slash). A STRICT SUBSET of the
+ * server's `INTERNAL_LINK_REGEX`:
+ *   - `<space>` = `[^/]+`   (any non-slash segment, as in the server's group 4)
+ *   - `<slug>`  = `[a-zA-Z0-9-]+`  (as in the server's group 5 / `extractPageSlugId`)
+ * Anchored on both ends so a scheme/host, a trailing `#anchor`/`?query`, or any
+ * extra path segment fails to match.
+ */
+const INTERNAL_PAGE_PATH = /^\/s\/[^/]+\/p\/[a-zA-Z0-9-]+\/?$/;
+
+/**
+ * True iff `href` is the unambiguous, space-qualified, root-relative internal
+ * page path. Pure and side-effect-free. Non-string input returns false.
+ */
+export function isInternalPagePath(href: unknown): boolean {
+  return typeof href === "string" && INTERNAL_PAGE_PATH.test(href);
+}
+
+/**
+ * The native internal-link attribute overrides applied to a link mark whose href
+ * is an internal page path. Mirrors the manual JSON-patch form
+ * (`{internal:true, target:null, rel:null}`) and the editor-ext/file-importer
+ * precedent: internal links carry no `target`/`rel` (same-tab SPA navigation).
+ */
+const INTERNAL_LINK_ATTRS = { internal: true, target: null, rel: null } as const;
+
+/**
+ * In-place post-walk of a finished ProseMirror doc that promotes every
+ * unambiguously-internal link mark to its native internal form.
+ *
+ * A link that spans several text nodes (e.g. `[**bold** word](/s/x/p/abc)`)
+ * stores an equivalent link mark on EACH covered text node — and a covered node
+ * may carry nested marks (bold/italic) alongside the link. We therefore walk the
+ * entire tree and rewrite EVERY `link` mark whose href passes
+ * `isInternalPagePath`, so no covered segment is left external.
+ *
+ * Pure of external effects and IDEMPOTENT: re-running on an already-marked doc
+ * leaves it unchanged (an internal-path href always maps to the same attrs).
+ * External links are never touched — their `target:_blank`/`rel:noopener…` stay.
+ *
+ * The doc is mutated in place (the converter owns the freshly-built doc and
+ * returns it directly); the same node reference is returned for convenience.
+ */
+export function markInternalLinks<T>(node: T): T {
+  walk(node);
+  return node;
+}
+
+function walk(node: any): void {
+  if (!node || typeof node !== "object") return;
+  if (Array.isArray(node)) {
+    for (const child of node) walk(child);
+    return;
+  }
+  if (Array.isArray(node.marks)) {
+    for (const mark of node.marks) {
+      if (
+        mark &&
+        mark.type === "link" &&
+        mark.attrs &&
+        isInternalPagePath(mark.attrs.href)
+      ) {
+        mark.attrs = { ...mark.attrs, ...INTERNAL_LINK_ATTRS };
+      }
+    }
+  }
+  if (Array.isArray(node.content)) walk(node.content);
+}

packages/prosemirror-markdown/src/lib/markdown-to-prosemirror.ts

@@ -12,6 +12,7 @@ import { parseHtmlDocument, generateJsonWith } from "./dom-parser.js";
 import type { TokenizerExtension, RendererExtension } from "marked";
 import { docmostExtensions } from "./docmost-schema.js";
 import { parseAttachedComment } from "./attached-comment.js";
+import { markInternalLinks } from "./internal-links.js";
 import { splitFootnoteParagraphs } from "./footnote.js";
 import {
   decodeInlineMathLatex,
@@ -1094,7 +1095,12 @@ export function markdownToProseMirrorSync(markdownContent: string): any {
   const withFootnotes = assembleFootnotes(withAttrs);
   const bridged = bridgeTaskLists(withFootnotes);
   const doc = generateJsonWith(bridged, docmostExtensions);
-  return stripEmptyParagraphs(doc);
+  // Promote unambiguously-internal wiki-page links (`[t](/s/<space>/p/<slug>)`)
+  // to their native internal form (`internal:true, target:null, rel:null`) so
+  // they get same-tab SPA navigation, hover-preview, and backlink participation.
+  // Every markdown import path funnels through here, so all of them are fixed at
+  // once; external links are left untouched (#522).
+  return markInternalLinks(stripEmptyParagraphs(doc));
 }
 
 /**

packages/prosemirror-markdown/test/internal-links.test.ts

@@ -0,0 +1,345 @@
+import { describe, expect, it } from 'vitest';
+import {
+  isInternalPagePath,
+  markInternalLinks,
+} from '../src/lib/internal-links.js';
+import { markdownToProseMirrorSync } from '../src/lib/markdown-to-prosemirror.js';
+import {
+  canonicalizeContent,
+  docsCanonicallyEqual,
+} from '../src/lib/canonicalize.js';
+
+// A LOCAL, illustrative copy of the server's INTERNAL_LINK_REGEX
+// (apps/server/src/integrations/export/utils.ts) used only to document the
+// subset boundary WITHIN this package (this layer cannot import the server).
+// It is NOT the cross-package drift guard: a hand copy can silently go stale if
+// the server regex is later narrowed. The real, mechanical drift guard lives in
+// the top layer, `apps/server/src/integrations/export/internal-link-parity.spec.ts`,
+// which imports BOTH the LIVE server `INTERNAL_LINK_REGEX` and the LIVE
+// `isInternalPagePath` and reddens if the client ever ceases to be a subset.
+const SERVER_INTERNAL_LINK_REGEX =
+  /^(https?:\/\/)?([^\/]+)?(\/s\/([^\/]+)\/)?p\/([a-zA-Z0-9-]+)\/?$/;
+
+// Walk a doc collecting every link mark (across all text nodes).
+const linkMarks = (doc: any): any[] => {
+  const out: any[] = [];
+  const walk = (n: any): void => {
+    if (!n || typeof n !== 'object') return;
+    if (Array.isArray(n)) return n.forEach(walk);
+    if (Array.isArray(n.marks))
+      for (const m of n.marks) if (m?.type === 'link') out.push(m);
+    if (Array.isArray(n.content)) walk(n.content);
+  };
+  walk(doc);
+  return out;
+};
+
+describe('isInternalPagePath', () => {
+  const ACCEPT = [
+    '/s/eng/p/abc123',
+    '/s/eng/p/abc123/', // trailing slash
+    '/s/My-Space/p/A-b-C-9', // hyphens + mixed case in both segments
+    '/s/x/p/z',
+  ];
+  const REJECT = [
+    'https://example.com/p/x', // scheme + host
+    'http://host/s/eng/p/abc', // scheme + host, even on the internal shape
+    '//host/s/eng/p/abc', // protocol-relative host
+    '/p/abc123', // space-less form (server does NOT backlink it)
+    'p/abc123', // relative
+    's/eng/p/abc123', // missing leading slash
+    '/s/eng/p/abc#section', // anchor
+    '/s/eng/p/abc?q=1', // query
+    '/s/eng/p/abc/extra', // extra segment
+    '/s//p/abc', // empty space segment
+    '/s/eng/p/', // empty slug
+    '/s/eng/p', // no slug at all
+    '/api/pages/abc', // other internal route
+    '/s/eng/x/abc', // wrong middle segment
+    '/s/a/b/p/abc', // space contains slash -> extra segment
+    '',
+  ];
+  // Structurally VALID `/s/<space>/p/<slug>` shapes whose ONLY defect is a
+  // forbidden character in the slug segment. These pin the slug charset
+  // `[a-zA-Z0-9-]` itself (not just the surrounding structure): drop them and a
+  // mutation widening the charset (e.g. adding `.`) survives the suite. Each is
+  // ALSO rejected by the server regex — documenting that the subset boundary
+  // holds precisely on the charset, not just on structure.
+  const REJECT_SLUG_CHARSET = [
+    '/s/eng/p/abc.def', // dot
+    '/s/eng/p/abc_def', // underscore
+    '/s/eng/p/abc%20', // percent-encoded space
+    '/s/eng/p/abc~x', // tilde
+  ];
+
+  it('accepts the space-qualified root-relative page path (+trailing slash)', () => {
+    for (const href of ACCEPT)
+      expect(isInternalPagePath(href), href).toBe(true);
+  });
+
+  it('rejects external URLs, ambiguous, and non-page forms', () => {
+    for (const href of REJECT)
+      expect(isInternalPagePath(href), href).toBe(false);
+  });
+
+  it('rejects a correctly-shaped path with a forbidden slug character', () => {
+    // Pins the slug charset itself: a mutation widening `[a-zA-Z0-9-]` reddens.
+    for (const href of REJECT_SLUG_CHARSET)
+      expect(isInternalPagePath(href), href).toBe(false);
+  });
+
+  it('the forbidden-slug-char paths are rejected by the server regex too', () => {
+    // The subset boundary holds on the charset, not just the structure: none of
+    // these match the server regex, so the client must not accept them either.
+    for (const href of REJECT_SLUG_CHARSET)
+      expect(SERVER_INTERNAL_LINK_REGEX.test(href), href).toBe(false);
+  });
+
+  it('rejects non-string input (fail-toward-external)', () => {
+    for (const v of [null, undefined, 123, {}, [], true])
+      expect(isInternalPagePath(v as unknown)).toBe(false);
+  });
+
+  it('is a STRICT SUBSET of the server INTERNAL_LINK_REGEX', () => {
+    // Every accepted href must also satisfy the server regex (so it is
+    // guaranteed backlink-able / export-rewritable).
+    for (const href of ACCEPT)
+      expect(SERVER_INTERNAL_LINK_REGEX.test(href), href).toBe(true);
+  });
+});
+
+describe('markInternalLinks', () => {
+  const linkNode = (text: string, href: string, extraMarks: any[] = []) => ({
+    type: 'text',
+    text,
+    marks: [
+      {
+        type: 'link',
+        attrs: {
+          href,
+          internal: null,
+          title: null,
+          target: '_blank',
+          rel: 'noopener noreferrer nofollow',
+          class: null,
+        },
+      },
+      ...extraMarks,
+    ],
+  });
+
+  it('marks an internal-path link internal:true, target:null, rel:null', () => {
+    const doc = {
+      type: 'doc',
+      content: [
+        { type: 'paragraph', content: [linkNode('hi', '/s/eng/p/abc123')] },
+      ],
+    };
+    markInternalLinks(doc);
+    const m = linkMarks(doc)[0];
+    expect(m.attrs.internal).toBe(true);
+    expect(m.attrs.target).toBeNull();
+    expect(m.attrs.rel).toBeNull();
+    expect(m.attrs.href).toBe('/s/eng/p/abc123'); // href untouched
+  });
+
+  it('leaves external links untouched', () => {
+    const doc = {
+      type: 'doc',
+      content: [
+        {
+          type: 'paragraph',
+          content: [linkNode('ext', 'https://example.com/p/x')],
+        },
+      ],
+    };
+    markInternalLinks(doc);
+    const m = linkMarks(doc)[0];
+    expect(m.attrs.internal).toBeNull();
+    expect(m.attrs.target).toBe('_blank');
+    expect(m.attrs.rel).toBe('noopener noreferrer nofollow');
+  });
+
+  it('marks EVERY text node a multi-node link spans (incl. nested bold/italic)', () => {
+    // A link `[**bold** plain *ital*](/s/x/p/abc)` stores the link mark on each
+    // covered text node; some also carry bold/italic. All must become internal.
+    const bold = { type: 'bold' };
+    const italic = { type: 'italic' };
+    const doc = {
+      type: 'doc',
+      content: [
+        {
+          type: 'paragraph',
+          content: [
+            linkNode('bold', '/s/x/p/abc', [bold]),
+            linkNode(' plain ', '/s/x/p/abc'),
+            linkNode('ital', '/s/x/p/abc', [italic]),
+          ],
+        },
+      ],
+    };
+    markInternalLinks(doc);
+    const marks = linkMarks(doc);
+    expect(marks).toHaveLength(3);
+    for (const m of marks) {
+      expect(m.attrs.internal).toBe(true);
+      expect(m.attrs.target).toBeNull();
+      expect(m.attrs.rel).toBeNull();
+    }
+    // Nested bold/italic marks are preserved alongside the promoted link.
+    const nested = doc.content[0].content;
+    expect(nested[0].marks.some((m: any) => m.type === 'bold')).toBe(true);
+    expect(nested[2].marks.some((m: any) => m.type === 'italic')).toBe(true);
+  });
+
+  it('is idempotent', () => {
+    const doc = {
+      type: 'doc',
+      content: [
+        { type: 'paragraph', content: [linkNode('hi', '/s/eng/p/abc')] },
+      ],
+    };
+    markInternalLinks(doc);
+    const once = JSON.stringify(doc);
+    markInternalLinks(doc);
+    expect(JSON.stringify(doc)).toBe(once);
+  });
+});
+
+describe('markdown import (full converter) — #522 acceptance', () => {
+  it('promotes an internal link and leaves an external one external', () => {
+    const doc = markdownToProseMirrorSync(
+      '[t](/s/eng/p/abc123) and [ext](https://example.com/p/x)',
+    );
+    const marks = linkMarks(doc);
+    const internal = marks.find((m) => m.attrs.href === '/s/eng/p/abc123');
+    const external = marks.find(
+      (m) => m.attrs.href === 'https://example.com/p/x',
+    );
+    expect(internal.attrs).toMatchObject({
+      internal: true,
+      target: null,
+      rel: null,
+    });
+    expect(external.attrs).toMatchObject({
+      internal: null,
+      target: '_blank',
+      rel: 'noopener noreferrer nofollow',
+    });
+  });
+
+  it('does NOT promote the space-less /p/<slug> form (documented limit)', () => {
+    const doc = markdownToProseMirrorSync('[t](/p/abc123)');
+    const m = linkMarks(doc)[0];
+    expect(m.attrs.internal).toBeNull();
+    expect(m.attrs.target).toBe('_blank');
+  });
+
+  it('the promoted link is backlink-extractable (matches server regex + true)', () => {
+    // Mirrors extractInternalLinkSlugIds: internal flag AND server-regex match.
+    const doc = markdownToProseMirrorSync('[t](/s/eng/p/my-page-abc123)');
+    const m = linkMarks(doc)[0];
+    expect(m.attrs.internal).toBe(true);
+    const match = m.attrs.href.match(SERVER_INTERNAL_LINK_REGEX);
+    expect(match).not.toBeNull();
+    // group 5 is the slug segment the server feeds to extractPageSlugId.
+    expect(match![5]).toBe('my-page-abc123');
+  });
+
+  it('comment-body markdown gets the same treatment (shared converter path)', () => {
+    // Comment bodies go through the same markdownToProseMirrorSync; a spot-check
+    // that the shared path (not a comment-only branch) does the promotion.
+    const doc = markdownToProseMirrorSync('see [here](/s/team/p/xyz789)');
+    const m = linkMarks(doc).find((x) => x.attrs.href === '/s/team/p/xyz789');
+    expect(m.attrs.internal).toBe(true);
+  });
+});
+
+describe('canonicalize — internal:false ≡ external (#522 §11)', () => {
+  it('drops editor-authored internal:false so external stays canonically equal', () => {
+    // Editor stores external links with internal:false; import leaves it absent.
+    const stored = {
+      type: 'doc',
+      content: [
+        {
+          type: 'paragraph',
+          content: [
+            {
+              type: 'text',
+              text: 'x',
+              marks: [
+                {
+                  type: 'link',
+                  attrs: {
+                    href: 'https://example.com',
+                    internal: false,
+                    target: '_blank',
+                    rel: 'noopener noreferrer nofollow',
+                  },
+                },
+              ],
+            },
+          ],
+        },
+      ],
+    };
+    const reimport = {
+      type: 'doc',
+      content: [
+        {
+          type: 'paragraph',
+          content: [
+            {
+              type: 'text',
+              text: 'x',
+              marks: [
+                {
+                  type: 'link',
+                  attrs: {
+                    href: 'https://example.com',
+                    internal: null, // import default
+                    target: '_blank',
+                    rel: 'noopener noreferrer nofollow',
+                  },
+                },
+              ],
+            },
+          ],
+        },
+      ],
+    };
+    expect(docsCanonicallyEqual(stored, reimport)).toBe(true);
+    // internal:false / target / rel all drop as defaults; only the non-default
+    // href survives.
+    const canon = canonicalizeContent(stored);
+    const mark = canon.content[0].content[0].marks[0];
+    expect(mark.attrs).toEqual({ href: 'https://example.com' });
+  });
+
+  it('keeps internal:true through canonicalization (non-default, load-bearing)', () => {
+    const doc = {
+      type: 'doc',
+      content: [
+        {
+          type: 'paragraph',
+          content: [
+            {
+              type: 'text',
+              text: 'x',
+              marks: [
+                {
+                  type: 'link',
+                  attrs: { href: '/s/x/p/abc', internal: true },
+                },
+              ],
+            },
+          ],
+        },
+      ],
+    };
+    const canon = canonicalizeContent(doc);
+    const mark = canon.content[0].content[0].marks[0];
+    expect(mark.attrs.internal).toBe(true);
+    expect(mark.attrs.href).toBe('/s/x/p/abc');
+  });
+});