diff --git a/pov-doc/SKILL.md b/pov-doc/SKILL.md
index ac91673..46dc724 100644
--- a/pov-doc/SKILL.md
+++ b/pov-doc/SKILL.md
@@ -66,8 +66,8 @@ The output document must follow this structure. Each section is required unless
 
 ### Header
 - Document title: `# **Verkada Proof of Concept Criteria**`
-- Customer logo placeholder: `<!-- Image goes here: [Customer Name] logo -->`
 - Verkada logo placeholder: `<!-- Image goes here: Verkada logo -->`
+- Customer logo placeholder: `<!-- Image goes here: [Customer Name] logo -->`
 - Customer mission statement (if known), formatted as a blockquote: `*"MISSION STATEMENT HERE"*`
 - Table of contents with anchor links to each numbered section. Each entry MUST be on its own line, like this:
   ```
diff --git a/pov-doc/examples/City of El Paso POC Criteria.md b/pov-doc/examples/City of El Paso POC Criteria.md
index f955c4b..4eb6be6 100644
--- a/pov-doc/examples/City of El Paso POC Criteria.md	
+++ b/pov-doc/examples/City of El Paso POC Criteria.md	
@@ -1,9 +1,9 @@
 # **Verkada Proof of Concept Criteria**
 
-<!-- Image goes here: customer logo -->
-
 # **Verkada logo goes here**
 
+<!-- Image goes here: customer logo -->
+
 # The mission of the City of El Paso is to deliver exceptional services to support a high quality of life and place for our community.
 
 ## **Table of Contents**
diff --git a/pov-doc/scripts/render_poc.py b/pov-doc/scripts/render_poc.py
index f579265..90e20e3 100644
--- a/pov-doc/scripts/render_poc.py
+++ b/pov-doc/scripts/render_poc.py
@@ -141,6 +141,13 @@ def process_markdown(
             md_content,
         )
 
+    # Insert page break before Table of Contents
+    md_content = re.sub(
+        r'(#{1,3} \*\*Table of Contents\*\*)',
+        r'\\newpage\n\n\1',
+        md_content,
+    )
+
     return md_content
 
 
@@ -228,6 +235,32 @@ def remove_heading_bookmarks(odt_path: str) -> None:
     os.replace(tmp_path, odt_path)
 
 
+def fix_image_frames(odt_path: str) -> None:
+    """Set all draw:frame elements in content.xml to inline (as-char) so text doesn't wrap around images."""
+    import xml.etree.ElementTree as ET
+
+    draw_ns = "urn:oasis:names:tc:opendocument:xmlns:drawing:1.0"
+    text_ns = "urn:oasis:names:tc:opendocument:xmlns:text:1.0"
+    svg_ns = "urn:oasis:names:tc:opendocument:xmlns:svg-compatible:1.0"
+
+    tmp_path = odt_path + ".tmp"
+    with zipfile.ZipFile(odt_path, "r") as zin:
+        with zipfile.ZipFile(tmp_path, "w", zipfile.ZIP_DEFLATED) as zout:
+            for item in zin.infolist():
+                data = zin.read(item.filename)
+                if item.filename == "content.xml":
+                    tree = ET.fromstring(data)
+                    for frame in tree.iter(f"{{{draw_ns}}}frame"):
+                        frame.set(f"{{{text_ns}}}anchor-type", "as-char")
+                        # Remove absolute positioning attrs that only apply to floating frames
+                        for attr in (f"{{{svg_ns}}}x", f"{{{svg_ns}}}y"):
+                            frame.attrib.pop(attr, None)
+                    data = ET.tostring(tree, encoding="unicode", xml_declaration=True).encode("utf-8")
+                zout.writestr(item, data)
+
+    os.replace(tmp_path, odt_path)
+
+
 def remove_header_images(odt_path: str) -> None:
     """Remove draw:frame elements from the header in styles.xml to strip reference doc logos."""
     ns = _odt_ns()
@@ -334,6 +367,7 @@ def main():
         if success:
             remove_heading_bookmarks(output_path)
             remove_header_images(output_path)
+            fix_image_frames(output_path)
             out_size = os.path.getsize(output_path)
             print(f"Done: {output_path} ({out_size:,} bytes)")
         else: