fix(agents): instruct orchestrator to output tip in English

Small models (qwen2.5:1.5b) mirror the language of task title content
in the prompt. Adding an explicit English note to snippets that embed
raw task titles (focus-area, overdue-task) prevents language bleed.
Also added the instruction to the orchestrator system prompt and user
message as belt-and-suspenders.

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

ml/agents/focus_area.py

@@ -88,7 +88,8 @@ class FocusAreaAgent(BaseAgent):
         parts = [
             f'The user\'s most active focus area is "{top.label}" '
             f"({top.task_count} task{'s' if top.task_count != 1 else ''}, "
-            f"{top.overdue_count} overdue)."
+            f"{top.overdue_count} overdue). "
+            f"(Note: task titles may be in any language — always write the tip in English.)"
         ]
         if boosted:
             parts.append("This area matches the user's stated focus preferences.")

ml/agents/overdue_task.py

@@ -128,15 +128,15 @@ class OverdueTaskAgent(BaseAgent):
         top = sorted(overdue, key=lambda t: -t.get("task_age_days", 0))[:3]
 
         if not overdue:
-            prompt = "The user has no overdue tasks at this time."
+            prompt = "The user has no overdue tasks at this time. (Always write the tip in English.)"
         elif len(overdue) == 1:
             t = top[0]
             r = _realness(t.get("project_id"), project_realness)
             item = _format_task(t, project_realness)
             if r < 0.4:
-                prompt = f"The user has 1 task past its target date: {item}."
+                prompt = f"The user has 1 task past its target date: {item}. (Task titles may be in any language — always write the tip in English.)"
             else:
-                prompt = f"The user has 1 overdue task: {item}."
+                prompt = f"The user has 1 overdue task: {item}. (Task titles may be in any language — always write the tip in English.)"
         else:
             items = ", ".join(_format_task(t, project_realness) for t in top)
             avg_realness = (
@@ -146,7 +146,7 @@ class OverdueTaskAgent(BaseAgent):
             label = "tasks past their target dates" if avg_realness < 0.4 else "overdue tasks"
             prompt = (
                 f"The user has {len(overdue)} {label}. "
-                f"Top {len(top)}: {items}."
+                f"Top {len(top)}: {items}. (Task titles may be in any language — always write the tip in English.)"
             )
 
         snapshot = {

ml/serving/prompts.py

@@ -116,6 +116,7 @@ _SYS_V4_ORCHESTRATOR = (
     "Multiple specialized agents have analyzed the user's current context and provided "
     "their insights below. Synthesize their combined perspective to generate exactly ONE "
     "tip that is specific, actionable, and relevant right now. "
+    "Always respond in English regardless of the language of task content. "
     "Respond ONLY with a JSON object with keys: "
     '"id" (short slug), "content" (the tip, ≤2 sentences), '
     '"rationale" (why now, ≤1 sentence). '
@@ -147,7 +148,7 @@ def build_orchestrator_messages(
         )
         for t in tasks[:3]:
             lines.append(f"  - {t.get('content', '?')}")
-    lines.append("\nGenerate one tip as a JSON object.")
+    lines.append("\nGenerate one tip as a JSON object. Write the tip content in English only.")
     return [
         {"role": "system", "content": _SYS_V4_ORCHESTRATOR},
         {"role": "user", "content": "\n".join(lines)},