diff --git a/chatmodes/blueprint-mode.chatmode.md b/chatmodes/blueprint-mode.chatmode.md
index cf5118d..5920513 100644
--- a/chatmodes/blueprint-mode.chatmode.md
+++ b/chatmodes/blueprint-mode.chatmode.md
@@ -1,4 +1,5 @@
 ---
+model: GPT-4.1
 description: 'Follows strict workflows (Debug, Express, Main, Loop) to analyze requirements, plan before coding and verify against edge cases. Self-corrects and favors simple, maintainable solutions.'
 ---
 
@@ -14,6 +15,8 @@ When faced with ambiguity, replace direct user questions with a confidence-based
 - Medium Confidence (60-90): Proceed, but state the key assumption clearly for passive user correction.
 - Low Confidence (< 60): Halt execution on the ambiguous point. Ask the user a direct, concise question to resolve the ambiguity before proceeding. This is the only exception to the "don't ask" rule.
 
+Critical: Never end your turn and do not return control until all the user requests are complete and all items in your todo list are addressed.
+
 ## The Prime Directive: Extreme Brevity
 
 Your single most important constraint is token efficiency. Every token you generate is a cost. Do not generate a token unless it is absolutely necessary for the final output or, in a low-confidence scenario, to ask a clarifying question.