diff --git a/nanobot/agent/loop.py b/nanobot/agent/loop.py
index 8be8e51..5d3c492 100644
--- a/nanobot/agent/loop.py
+++ b/nanobot/agent/loop.py
@@ -427,12 +427,15 @@ class AgentLoop:
     def _save_turn(self, session: Session, messages: list[dict], skip: int) -> None:
         """Save new-turn messages into session, truncating large tool results."""
         from datetime import datetime
+        from nanobot.utils import helpers
         for m in messages[skip:]:
             entry = {k: v for k, v in m.items() if k != "reasoning_content"}
             if entry.get("role") == "tool" and isinstance(entry.get("content"), str):
                 content = entry["content"]
                 if len(content) > self._TOOL_RESULT_MAX_CHARS:
-                    entry["content"] = content[:self._TOOL_RESULT_MAX_CHARS] + "\n... (truncated)"
+                    entry["content"] = content[: self._TOOL_RESULT_MAX_CHARS] + "\n... (truncated)"
+            if entry.get("role") == "user":
+                entry["content"] = helpers.strip_base64_images(entry["content"])
             entry.setdefault("timestamp", datetime.now().isoformat())
             session.messages.append(entry)
         session.updated_at = datetime.now()
diff --git a/nanobot/utils/helpers.py b/nanobot/utils/helpers.py
index 62f80ac..c977473 100644
--- a/nanobot/utils/helpers.py
+++ b/nanobot/utils/helpers.py
@@ -2,6 +2,7 @@
 
 from pathlib import Path
 from datetime import datetime
+from typing import Any
 
 
 def ensure_dir(path: Path) -> Path:
@@ -78,3 +79,28 @@ def parse_session_key(key: str) -> tuple[str, str]:
     if len(parts) != 2:
         raise ValueError(f"Invalid session key: {key}")
     return parts[0], parts[1]
+
+def strip_base64_images(content: str | list[dict[str, Any]]) -> str | list[dict[str, Any]]:
+    """Strip base64 image data from message content, replacing with text placeholder."""
+    if not isinstance(content, list):
+        return content
+
+    new_content = []
+    for item in content:
+        if not isinstance(item, dict):
+            new_content.append(item)
+            continue
+
+        if item.get("type") == "image_url":
+            url = item.get("image_url", {}).get("url", "")
+            if url.startswith("data:image/") and ";base64," in url:
+                new_content.append({"type": "text", "text": "[image]"})
+                continue
+        new_content.append(item)
+
+    text_parts = [c["text"] for c in new_content if isinstance(c, dict) and c.get("type") == "text"]
+    if len(new_content) == 1 and not text_parts:
+        return new_content[0] if new_content else ""
+    if text_parts and len(new_content) == len(text_parts):
+        return "\n".join(text_parts)
+    return new_content
\ No newline at end of file