From f8a5df9d5a697e7739919fca66d73370c08df9ad Mon Sep 17 00:00:00 2001
From: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Date: Wed, 12 Feb 2025 19:53:41 +0000
Subject: [PATCH] feat: Add feedback validation and error handling

- Add FeedbackProcessingError for feedback handling
- Add validation for empty and long feedback messages
- Add test coverage for edge cases

Co-Authored-By: Joe Moura <joao@crewai.com>
---
 src/crewai/agents/crew_agent_executor.py | 18 ++++++++
 tests/agent_test.py                      | 52 +++++++++++++++++++++++-
 2 files changed, 68 insertions(+), 2 deletions(-)

diff --git a/src/crewai/agents/crew_agent_executor.py b/src/crewai/agents/crew_agent_executor.py
index 0ea229119..b4a246a63 100644
--- a/src/crewai/agents/crew_agent_executor.py
+++ b/src/crewai/agents/crew_agent_executor.py
@@ -18,6 +18,7 @@ from crewai.tools.base_tool import BaseTool
 from crewai.tools.tool_usage import ToolUsage, ToolUsageErrorException
 from crewai.utilities import I18N, Printer
 from crewai.utilities.constants import MAX_LLM_RETRY, TRAINING_DATA_FILE
+from crewai.utilities.exceptions.feedback_processing_exception import FeedbackProcessingError
 from crewai.utilities.exceptions.context_window_exceeding_exception import (
     LLMContextLengthExceededException,
 )
@@ -487,6 +488,23 @@ class CrewAgentExecutor(CrewAgentExecutorMixin):
         return CrewAgentParser(agent=self.agent).parse(answer)
 
     def _format_msg(self, prompt: str, role: str = "user") -> Dict[str, str]:
+        """Format a message with role and content.
+        
+        Args:
+            prompt (str): The message content
+            role (str): The message role (default: "user")
+            
+        Returns:
+            Dict[str, str]: Formatted message with role and content
+            
+        Raises:
+            FeedbackProcessingError: If prompt is empty or exceeds max length
+        """
+        if not prompt or not prompt.strip():
+            raise FeedbackProcessingError("Feedback message cannot be empty")
+        if len(prompt) > 8192:  # Standard context window size
+            raise FeedbackProcessingError("Feedback message exceeds maximum length")
+            
         prompt = prompt.rstrip()
         return {"role": role, "content": prompt}
 
diff --git a/tests/agent_test.py b/tests/agent_test.py
index e67a7454a..1593b7a60 100644
--- a/tests/agent_test.py
+++ b/tests/agent_test.py
@@ -2,13 +2,14 @@
 
 import os
 from unittest import mock
-from unittest.mock import patch
+from unittest.mock import patch, MagicMock
 
 import pytest
 
 from crewai import Agent, Crew, Task
-from crewai.agents.cache import CacheHandler
 from crewai.agents.crew_agent_executor import CrewAgentExecutor
+from crewai.utilities.exceptions.feedback_processing_exception import FeedbackProcessingError
+from crewai.agents.cache import CacheHandler
 from crewai.agents.parser import AgentAction, CrewAgentParser, OutputParserException
 from crewai.knowledge.source.base_knowledge_source import BaseKnowledgeSource
 from crewai.knowledge.source.string_knowledge_source import StringKnowledgeSource
@@ -1001,6 +1002,53 @@ def test_agent_human_input():
         assert mock_human_input.call_count == 2  # Should have asked for feedback twice
         assert output.strip().lower() == "hello"  # Final output should be 'Hello'
 
+        # Verify message format for human feedback
+        messages = agent.agent_executor.messages
+        feedback_messages = [m for m in messages if "Feedback:" in m.get("content", "")]
+        assert len(feedback_messages) == 2  # Two feedback messages
+        for msg in feedback_messages:
+            assert msg["role"] == "user"  # All feedback messages should have user role
+
+
+@pytest.fixture
+def mock_executor():
+    """Create a mock executor for testing."""
+    agent = Agent(
+        role="test role",
+        goal="test goal",
+        backstory="test backstory"
+    )
+    task = Task(
+        description="Test task",
+        expected_output="Test output",
+        human_input=True,
+        agent=agent
+    )
+    executor = CrewAgentExecutor(
+        agent=agent,
+        task=task,
+        llm=agent.llm,
+        crew=None,
+        prompt="",
+        max_iter=1,
+        tools=[],
+        tools_names=[],
+        stop_words=[],
+        tools_description="",
+        tools_handler=None
+    )
+    return executor
+
+def test_empty_feedback_handling(mock_executor):
+    """Test that empty feedback is properly handled."""
+    with pytest.raises(FeedbackProcessingError):
+        mock_executor._format_msg("")
+
+def test_long_feedback_handling(mock_executor):
+    """Test that very long feedback is properly handled."""
+    very_long_feedback = "x" * 10000
+    with pytest.raises(FeedbackProcessingError):
+        mock_executor._format_msg(very_long_feedback)
 
 def test_interpolate_inputs():
     agent = Agent(