diff --git a/surfsense_backend/app/tasks/chat/stream_new_chat.py b/surfsense_backend/app/tasks/chat/stream_new_chat.py
index 940538bff..23c3daee2 100644
--- a/surfsense_backend/app/tasks/chat/stream_new_chat.py
+++ b/surfsense_backend/app/tasks/chat/stream_new_chat.py
@@ -52,38 +52,41 @@ def format_mentioned_documents_as_context(documents: list[Document]) -> str:
     """Format mentioned documents as context for the agent."""
     if not documents:
         return ""
+    import json
 
-    context_parts = ["<mentioned_documents>"]
-    context_parts.append(
-        "The user has explicitly mentioned the following documents from their knowledge base. "
-        "These documents are directly relevant to the query and should be prioritized as primary sources."
-    )
-    for i, doc in enumerate(documents, 1):
-        # Prepare retriever-style structure
-        doc_metadata = doc.document_metadata if doc.document_metadata else {}
-        xml_doc = f"""
-        <document index='{i}'>
-            <document_id>{doc.id}</document_id>
-            <content><![CDATA[{doc.content}]]></content>
-            <score>1.0</score>
-            <chunks>
-                <chunk>
-                    <chunk_id>{doc.id}-full</chunk_id>
-                    <content><![CDATA[{doc.content}]]></content>
-                </chunk>
-            </chunks>
-            <document_info>
-                <id>{doc.id}</id>
-                <title><![CDATA[{doc.title}]]></title>
-                <document_type>{doc.document_type.value}</document_type>
-                <metadata><![CDATA[{json.dumps(doc_metadata)}]]></metadata>
-            </document_info>
-            <source>{doc.document_type.value}</source>
-        </document>"""
-        context_parts.append(xml_doc.strip())
-    context_parts.append("</mentioned_documents>")
-
-    return "\n".join(context_parts)
+    parts = []
+    for doc in documents:
+        metadata = doc.document_metadata or {}
+        chunks = (
+            [
+                {"chunk_id": c.id, "content": c.content}
+                for c in getattr(doc, "chunks", [])
+            ]
+            if hasattr(doc, "chunks") and doc.chunks
+            else [{"chunk_id": doc.id, "content": doc.content}]
+        )
+        metadata_json = json.dumps(metadata, ensure_ascii=False)
+        parts.append("<document>")
+        parts.append("<document_metadata>")
+        parts.append(f"  <document_id>{doc.id}</document_id>")
+        parts.append(f"  <document_type>{doc.document_type.value}</document_type>")
+        parts.append(f"  <title><![CDATA[{doc.title}]]></title>")
+        parts.append("  <url><![CDATA[]]></url>")
+        parts.append(f"  <metadata_json><![CDATA[{metadata_json}]]></metadata_json>")
+        parts.append("</document_metadata>")
+        parts.append("")
+        parts.append("<document_content>")
+        for ch in chunks:
+            ch_content = ch["content"]
+            ch_id = ch["chunk_id"]
+            if ch_id is None:
+                parts.append(f"  <chunk><![CDATA[{ch_content}]]></chunk>")
+            else:
+                parts.append(f"  <chunk id='{ch_id}'><![CDATA[{ch_content}]]></chunk>")
+        parts.append("</document_content>")
+        parts.append("</document>")
+        parts.append("")
+    return "\n".join(parts).strip()
 
 
 async def stream_new_chat(