fix: extract clean task title from natural language phrasings

- Item noun phrases (job item, work item) checked before "add X to Y"
  so the wrong " to " isn't captured in multi-clause sentences
- "please add X to [project], title" — comma after project = explicit title
- "work item to [project], title" — strips "to " filler then takes after comma
- Restructured function: explicit prefixes → item nouns → add-to pattern

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

app/intent.py

@@ -203,22 +203,48 @@ def extract_task_title(message: str) -> str:
     import re as _re
     msg = message.strip()
     lower = msg.lower()
-    if lower.startswith("add ") and " to " in lower:
+
-        parts = msg.rsplit(" to ", 1)
+    # Explicit prefixes
-        title = parts[0][4:].strip()
-        return title if title else msg
     for prefix in ("create task:", "add task:", "log task:", "new task:", "log this:"):
         if lower.startswith(prefix):
             return msg[len(prefix):].strip()
-    # "add [who] a job item [description]" — extract description after "job item"
+
-    m = _re.search(r'\bjob item\b\s*(.*)', lower)
+    # Item noun phrases — "add X a job/work item [description]"
+    # Check these before "add X to Y" so the wrong " to " isn't captured
+    for noun in (r'\bjob item\b', r'\bwork item\b', r'\bwork order\b', r'\bjob ticket\b'):
+        m = _re.search(noun + r'\s*(.*)', lower)
         if m:
             remainder = msg[m.start(1):].strip()
             for filler in ("that we need to ", "that is ", "which is ", "to "):
                 if remainder.lower().startswith(filler):
                     remainder = remainder[len(filler):]
+            # "project, title" leftover → take after the comma
+            if "," in remainder:
+                remainder = remainder[remainder.index(",") + 1:].strip()
             if remainder.strip():
                 return remainder.strip()
+
+    # "please add X to [project], title" or "add X to [project], title"
+    work_msg = msg[7:].strip() if lower.startswith("please ") else msg
+    work_lower = work_msg.lower()
+    if work_lower.startswith("add ") and " to " in work_lower:
+        to_idx = work_lower.rfind(" to ")
+        after_to = work_msg[to_idx + 4:]
+        if "," in after_to:
+            title_candidate = after_to[after_to.index(",") + 1:].strip()
+            if title_candidate:
+                return title_candidate
+        # No comma — what's between "add " and " to [project]" is the title
+        before_to = work_msg[4:to_idx].strip()
+        for noise in ("a work item", "an issue", "a task", "a job item", "a ticket",
+                      "an item", "a to-do", "a todo"):
+            if before_to.lower().startswith(noise):
+                remainder = before_to[len(noise):].strip()
+                if remainder:
+                    return remainder
+        if before_to:
+            return before_to
+
     return msg