diff --git a/src/primaite/agents/rllib.py b/src/primaite/agents/rllib.py
index 4149d02d..fb062f54 100644
--- a/src/primaite/agents/rllib.py
+++ b/src/primaite/agents/rllib.py
@@ -75,7 +75,7 @@ class RLlibAgent(AgentSessionABC):
 
         super().__init__(training_config_path, lay_down_config_path)
         if self._training_config.session_type == SessionType.EVAL:
-            msg = "Cannot evaluate an RLlib agent that hasn't been through trainig yet."
+            msg = "Cannot evaluate an RLlib agent that hasn't been through training yet."
             _LOGGER.critical(msg)
             raise RLlibAgentError(msg)
         if not self._training_config.agent_framework == AgentFramework.RLLIB:
@@ -194,6 +194,7 @@ class RLlibAgent(AgentSessionABC):
             self._plot_av_reward_per_episode(learning_session=True)
 
     def _unpack_saved_agent_into_eval(self) -> Path:
+        """Unpacks the pre-trained and saved RLlib agent so that it can be reloaded by Ray for eval."""
         agent_restore_path = self.evaluation_path / "agent_restore"
         if agent_restore_path.exists():
             shutil.rmtree(agent_restore_path)
@@ -248,6 +249,9 @@ class RLlibAgent(AgentSessionABC):
         if self._training_config.session_type is not SessionType.TRAIN:
             self._train_agent.stop()
             self._plot_av_reward_per_episode(learning_session=True)
+        # Perform a clean-up of the unpacked agent
+        if (self.evaluation_path / "agent_restore").exists():
+            shutil.rmtree((self.evaluation_path / "agent_restore"))
 
     def _get_latest_checkpoint(self) -> None:
         raise NotImplementedError