fix: preserve existing bio when using --skip-llm

Added get_existing_bio() to check spreadsheet before generating fallback bio.
Now --skip-llm will use existing bio from spreadsheet if available.

Author: jeremymanning

data/people.xlsx

@@ -659,6 +659,7 @@ \subsection*{Mentorship  (selected)}
 
 \textit{Graduate Advisees}:
 \begin{etaremune}
+  \item Theo Larson (Doctoral student; 2026 -- )
 \item Claudia Gonciulea (Doctoral student; 2025 -- )
 \item Paxton Fitzpatrick (Doctoral student; 2021 -- )
 \item Xinming Xu (Doctoral student; 2021 -- )

documents/JRM_CV.tex

@@ -192,6 +192,11 @@ <h3>sam haskel | undergrad</h3>
                     <h3>will lehman | undergrad</h3>
                     <p>Will is a '27 from Salt Lake City, Utah. He likes skiing, playing hockey, and working on new projects.</p>
                 </div>
+                <div class="person-card">
+                    <img src="images/people/theo_larson.png" alt="theo larson">
+                    <h3>theo larson | undergrad</h3>
+                    <p>Theo is a '29 from Brooklyn, New York and a prospective major in Economics and History. In his free time, he enjoys reading, playing board games, and taking walks with his dog Sunny.</p>
+                </div>
             </div>
         </section>
 

images/people/theo_larson.png

@@ -678,6 +678,23 @@ def member_exists_in_spreadsheet(
     return False, None
 
 
+def get_existing_bio(xlsx_path: Path, name: str) -> Optional[str]:
+    wb = openpyxl.load_workbook(xlsx_path)
+    sheet = wb["members"]
+
+    name_lower = name.lower()
+    for row in sheet.iter_rows(min_row=2, values_only=True):
+        if row[1] and str(row[1]).lower() == name_lower:
+            bio = row[4] if len(row) > 4 else None
+            wb.close()
+            if bio and len(str(bio)) > 10:
+                return str(bio)
+            return None
+
+    wb.close()
+    return None
+
+
 def find_alumni_entry(xlsx_path: Path, name: str) -> Optional[Dict[str, Any]]:
     """Find alumni entry across all alumni sheets. Returns dict with sheet name, row, and data."""
     wb = openpyxl.load_workbook(xlsx_path)
@@ -1024,6 +1041,12 @@ def onboard_member(
         print("  Searching git history for old bio...")
         bio = get_bio_from_git_history(xlsx_path, name, project_root)
 
+    if bio is None:
+        existing_bio = get_existing_bio(xlsx_path, name)
+        if existing_bio:
+            print(f"  Using existing bio from spreadsheet")
+            bio = existing_bio
+
     if not skip_llm:
         if bio:
             print("  Editing bio with LLM...")