Visualize chunk command, format citations to show chunk id

haiku_rag_slim/haiku/rag/app.py

@@ -274,6 +274,38 @@ class HaikuRAGApp:
             for result in results:
                 self._rich_print_search_result(result)
 
+    async def visualize_chunk(self, chunk_id: str):
+        """Display visual grounding images for a chunk."""
+        from textual_image.renderable import Image as RichImage
+
+        async with HaikuRAG(db_path=self.db_path, config=self.config) as self.client:
+            chunk = await self.client.chunk_repository.get_by_id(chunk_id)
+            if not chunk:
+                self.console.print(f"[red]Chunk with id {chunk_id} not found.[/red]")
+                return
+
+            images = await self.client.visualize_chunk(chunk)
+            if not images:
+                self.console.print(
+                    "[yellow]No visual grounding available for this chunk.[/yellow]"
+                )
+                self.console.print(
+                    "This may be because the document was converted without page images."
+                )
+                return
+
+            self.console.print(f"[bold]Visual grounding for chunk {chunk_id}[/bold]")
+            if chunk.document_uri:
+                self.console.print(
+                    f"[repr.attrib_name]document[/repr.attrib_name]: {chunk.document_uri}"
+                )
+
+            for i, img in enumerate(images):
+                self.console.print(
+                    f"\n[bold cyan]Page {i + 1}/{len(images)}[/bold cyan]"
+                )
+                self.console.print(RichImage(img))
+
     async def ask(
         self,
         question: str,
@@ -624,11 +656,13 @@ class HaikuRAGApp:
         content = Markdown(result.content)
         self.console.print(
             f"[repr.attrib_name]document_id[/repr.attrib_name]: {result.document_id} "
+            f"[repr.attrib_name]chunk_id[/repr.attrib_name]: {result.chunk_id} "
             f"[repr.attrib_name]score[/repr.attrib_name]: {result.score:.4f}"
         )
         if result.document_uri:
-            self.console.print("[repr.attrib_name]document uri[/repr.attrib_name]:")
-            self.console.print(result.document_uri)
+            self.console.print(
+                f"[repr.attrib_name]document uri[/repr.attrib_name]: {result.document_uri}"
+            )
         if result.document_title:
             self.console.print("[repr.attrib_name]document title[/repr.attrib_name]:")
             self.console.print(result.document_title)

haiku_rag_slim/haiku/rag/cli.py

@@ -261,6 +261,21 @@ def search(
     asyncio.run(app.search(query=query, limit=limit, filter=filter))
 
 
+@cli.command("visualize", help="Show visual grounding for a chunk")
+def visualize(
+    chunk_id: str = typer.Argument(
+        help="The ID of the chunk to visualize",
+    ),
+    db: Path | None = typer.Option(
+        None,
+        "--db",
+        help="Path to the LanceDB database file",
+    ),
+):
+    app = create_app(db)
+    asyncio.run(app.visualize_chunk(chunk_id=chunk_id))
+
+
 @cli.command("ask", help="Ask a question using the QA agent")
 def ask(
     question: str = typer.Argument(

haiku_rag_slim/haiku/rag/graph/common/models.py

@@ -31,6 +31,8 @@ class ResearchPlan(BaseModel):
 class Citation(BaseModel):
     """Resolved citation with full metadata for display/visual grounding."""
 
+    document_id: str
+    chunk_id: str
     document_uri: str
     document_title: str | None = None
     page_numbers: list[int] = Field(default_factory=list)
@@ -75,6 +77,8 @@ def resolve_citations(
             continue
         citations.append(
             Citation(
+                document_id=r.document_id or "",
+                chunk_id=chunk_id,
                 document_uri=r.document_uri or "",
                 document_title=r.document_title,
                 page_numbers=r.page_numbers,

haiku_rag_slim/haiku/rag/utils.py

@@ -280,8 +280,11 @@ def format_citations(citations: "list[Citation]") -> str:
         return ""
     lines = ["## Citations\n"]
     for c in citations:
-        # Build citation header
-        parts = [f"- **{c.document_uri}**"]
+        # Build citation header with document_id and chunk_id
+        parts = [
+            f"- document_id: `{c.document_id}` chunk_id: `{c.chunk_id}` "
+            f"uri: **{c.document_uri}**"
+        ]
         if c.document_title:
             parts.append(f' - "{c.document_title}"')
         location_parts = []