diff --git a/autogpts/autogpt/autogpt/agents/base.py b/autogpts/autogpt/autogpt/agents/base.py
index 774070e8c6e9..f71d8c7a8bae 100644
--- a/autogpts/autogpt/autogpt/agents/base.py
+++ b/autogpts/autogpt/autogpt/agents/base.py
@@ -40,7 +40,11 @@
 from autogpt.core.runner.client_lib.logging.helpers import dump_prompt
 from autogpt.file_storage.base import FileStorage
 from autogpt.llm.providers.openai import get_openai_command_specs
-from autogpt.models.action_history import ActionResult, EpisodicActionHistory
+from autogpt.models.action_history import (
+    ActionResult,
+    ActionSuccessResult,
+    EpisodicActionHistory,
+)
 from autogpt.prompts.prompt import DEFAULT_TRIGGERING_PROMPT
 
 logger = logging.getLogger(__name__)
@@ -173,6 +177,14 @@ def __init__(
         self.legacy_config = legacy_config
         """LEGACY: Monolithic application configuration."""
 
+        # In case the agent is resumed, cursor is set to the last episode
+        if self.event_history:
+            # To prevent errors, when the last action is "finish", we register a result
+            # And move cursor to the next action
+            if self.event_history.current_episode.action.name == "finish":
+                self.event_history.register_result(ActionSuccessResult())
+                self.event_history.cursor = len(self.event_history)
+
         self.llm_provider = llm_provider
 
         self.prompt_strategy = prompt_strategy
diff --git a/autogpts/autogpt/autogpt/app/agent_protocol_server.py b/autogpts/autogpt/autogpt/app/agent_protocol_server.py
index 9cc0603b2d3f..5118782d157e 100644
--- a/autogpts/autogpt/autogpt/app/agent_protocol_server.py
+++ b/autogpts/autogpt/autogpt/app/agent_protocol_server.py
@@ -311,12 +311,16 @@ async def execute_step(self, task_id: str, step_request: StepRequestBody) -> Ste
                         "name": execute_command,
                         "args": execute_command_args,
                         "result": (
-                            orjson.loads(execute_result.json())
-                            if not isinstance(execute_result, ActionErrorResult)
-                            else {
-                                "error": str(execute_result.error),
-                                "reason": execute_result.reason,
-                            }
+                            ""
+                            if execute_result is None
+                            else (
+                                orjson.loads(execute_result.json())
+                                if not isinstance(execute_result, ActionErrorResult)
+                                else {
+                                    "error": str(execute_result.error),
+                                    "reason": execute_result.reason,
+                                }
+                            )
                         ),
                     },
                 }
diff --git a/autogpts/autogpt/autogpt/app/main.py b/autogpts/autogpt/autogpt/app/main.py
index 50c9bfff26a9..ad04bad8fb21 100644
--- a/autogpts/autogpt/autogpt/app/main.py
+++ b/autogpts/autogpt/autogpt/app/main.py
@@ -190,16 +190,14 @@ async def run_auto_gpt(
     # Resume an Existing Agent #
     ############################
     if load_existing_agent:
-        agent_state = agent_manager.load_agent_state(load_existing_agent)
+        agent_state = None
         while True:
             answer = await clean_input(config, "Resume? [Y/n]")
-            if answer.lower() == "y":
+            if answer == "" or answer.lower() == "y":
+                agent_state = agent_manager.load_agent_state(load_existing_agent)
                 break
             elif answer.lower() == "n":
-                agent_state = None
                 break
-            else:
-                print("Please respond with 'y' or 'n'")
 
     if agent_state:
         agent = configure_agent_with_state(