diff --git a/backend/app/core/ai_provider.py b/backend/app/core/ai_provider.py
index cea16269..cb3f7178 100644
--- a/backend/app/core/ai_provider.py
+++ b/backend/app/core/ai_provider.py
@@ -5,10 +5,13 @@ Supports Gemini (google-genai) and Anthropic (anthropic) as interchangeable
 backends for JSON generation used by the AI Flow Builder.
 """
 
+import logging
 from abc import ABC, abstractmethod
 
 from app.core.config import settings
 
+logger = logging.getLogger(__name__)
+
 
 class AIProvider(ABC):
     """Abstract base class for AI providers."""
@@ -74,6 +77,16 @@ class GeminiProvider(AIProvider):
             config=config,
         )
 
+        # Log finish reason to detect truncation
+        if response.candidates:
+            finish_reason = getattr(response.candidates[0], "finish_reason", None)
+            logger.info("Gemini finish_reason=%s model=%s", finish_reason, self._model)
+            if str(finish_reason) == "MAX_TOKENS":
+                logger.warning(
+                    "Gemini output truncated (MAX_TOKENS). max_output_tokens=%d",
+                    max_tokens,
+                )
+
         text = response.text or ""
         input_tokens = getattr(response.usage_metadata, "prompt_token_count", 0) or 0
         output_tokens = (
diff --git a/backend/app/core/ai_tree_generator_service.py b/backend/app/core/ai_tree_generator_service.py
index 7a562d1c..bf560874 100644
--- a/backend/app/core/ai_tree_generator_service.py
+++ b/backend/app/core/ai_tree_generator_service.py
@@ -154,15 +154,23 @@ async def scaffold_branches(
     raw_text, input_tokens, output_tokens = await provider.generate_json(
         system_prompt=SCAFFOLD_SYSTEM_PROMPT,
         messages=[{"role": "user", "content": user_message}],
-        max_tokens=1024,
+        max_tokens=2048,
     )
 
+    logger.info(
+        "scaffold raw response (tokens in=%d out=%d, len=%d): %s",
+        input_tokens,
+        output_tokens,
+        len(raw_text),
+        raw_text[:500],
+    )
     raw_text = _strip_markdown_fences(raw_text)
     cost = _estimate_cost(input_tokens, output_tokens)
 
     try:
         data = json.loads(raw_text)
     except json.JSONDecodeError as e:
+        logger.error("scaffold JSON parse failed. Full text (%d chars): %s", len(raw_text), raw_text)
         raise ValueError(f"AI returned invalid JSON: {e}")
 
     branches = data.get("branches", [])
@@ -224,6 +232,10 @@ async def generate_branch_detail(
         try:
             branch_tree = json.loads(raw_text)
         except json.JSONDecodeError as e:
+            logger.error(
+                "branch_detail attempt=%d JSON parse failed (%d chars): %s",
+                attempt, len(raw_text), raw_text[:500],
+            )
             if attempt < 2:
                 messages.append({"role": "assistant", "content": raw_text})
                 messages.append({