diff --git a/README.md b/README.md
index 9f02977..73d5c22 100644
--- a/README.md
+++ b/README.md
@@ -24,7 +24,8 @@ limitations under the License.
 </p>
 
 <h3 align="center">
-  <p>🤗 smolagents - a smol library to build great agents!</p>
+  <img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/smolagents/mascot.png" alt="Hugging Face mascot as James Bond" width=50px>
+  <p>smolagents - a smol library to build great agents!</p>
 </h3>
 
 `smolagents` is a library that enables you to run powerful agents in a few lines of code. It offers:
diff --git a/src/smolagents/agents.py b/src/smolagents/agents.py
index cfa8a6f..6a2bcde 100644
--- a/src/smolagents/agents.py
+++ b/src/smolagents/agents.py
@@ -75,12 +75,12 @@ class ToolCall:
     id: str
 
 
-class AgentStep:
+class AgentStepLog:
     pass
 
 
 @dataclass
-class ActionStep(AgentStep):
+class ActionStep(AgentStepLog):
     agent_memory: List[Dict[str, str]] | None = None
     tool_calls: List[ToolCall] | None = None
     start_time: float | None = None
@@ -90,22 +90,24 @@ class ActionStep(AgentStep):
     duration: float | None = None
     llm_output: str | None = None
     observations: str | None = None
+    observations_images: List[Image]
     action_output: Any = None
 
 
 @dataclass
-class PlanningStep(AgentStep):
+class PlanningStep(AgentStepLog):
     plan: str
     facts: str
 
 
 @dataclass
-class TaskStep(AgentStep):
+class TaskStep(AgentStepLog):
     task: str
+    task_images: List[Image]
 
 
 @dataclass
-class SystemPromptStep(AgentStep):
+class SystemPromptStep(AgentStepLog):
     system_prompt: str