fix(agents-api): Retry failed integration executions (#874)

<!-- ELLIPSIS_HIDDEN -->



> [!IMPORTANT]
> Introduces `IntegrationExecutionException` for integration execution
errors and updates retry logic to handle it.
> 
>   - **Behavior**:
> - Introduces `IntegrationExecutionException` in `tools.py` for errors
during integration execution.
> - Updates `execute_integration()` in `execute_integration.py` to raise
`IntegrationExecutionException` on integration service errors.
>   - **Error Handling**:
> - Adds `IntegrationExecutionException` to `RETRYABLE_ERROR_TYPES` in
`tasks.py` to allow retries for integration execution errors.
> 
> <sup>This description was created by </sup>[<img alt="Ellipsis"
src="https://img.shields.io/badge/Ellipsis-blue?color=175173">](https://www.ellipsis.dev?ref=julep-ai%2Fjulep&utm_source=github&utm_medium=referral)<sup>
for 11edfc0. It will automatically
update as commits are pushed.</sup>

<!-- ELLIPSIS_HIDDEN -->

---------

Co-authored-by: HamadaSalhab <[email protected]>

agents-api/agents_api/activities/execute_integration.py

@@ -5,6 +5,7 @@
 
 from ..autogen.openapi_model import BaseIntegrationDef
 from ..clients import integrations
+from ..common.exceptions.tools import IntegrationExecutionException
 from ..common.protocol.tasks import StepContext
 from ..common.storage_handler import auto_blob_store
 from ..env import testing
@@ -53,13 +54,21 @@ async def execute_integration(
             "error" in integration_service_response
             and integration_service_response["error"]
         ):
-            raise Exception(integration_service_response["error"])
+            raise IntegrationExecutionException(
+                integration=integration,
+                error=integration_service_response["error"],
+            )
 
         return integration_service_response
 
     except BaseException as e:
         if activity.in_activity():
-            activity.logger.error(f"Error in execute_integration: {e}")
+            integration_str = integration.provider + (
+                "." + integration.method if integration.method else ""
+            )
+            activity.logger.error(
+                f"Error in execute_integration {integration_str}: {e}"
+            )
 
         raise
 

agents-api/agents_api/common/exceptions/tasks.py

@@ -26,6 +26,8 @@
 import temporalio.exceptions
 from tenacity import RetryError
 
+from .tools import IntegrationExecutionException
+
 # 🚫 The "No Second Chances" Club - errors that we won't retry
 # Because sometimes, no means no!
 NON_RETRYABLE_ERROR_TYPES = (
@@ -134,6 +136,9 @@
     #
     # Tenacity exceptions (retry when retrying goes wrong lol)
     RetryError,
+    #
+    # Tools-related exceptions (when tools have a bad day)
+    IntegrationExecutionException,
 )
 
 # HTTP status codes that say "maybe try again later?"

agents-api/agents_api/common/exceptions/tools.py

@@ -0,0 +1,25 @@
+"""
+Defines tools-related exceptions for the agents API.
+"""
+
+from ...autogen.openapi_model import BaseIntegrationDef
+from . import BaseCommonException
+
+
+class BaseToolsException(BaseCommonException):
+    """Base exception for tools-related errors."""
+
+    pass
+
+
+class IntegrationExecutionException(BaseToolsException):
+    """Exception raised when an error occurs during an integration execution."""
+
+    def __init__(self, integration: BaseIntegrationDef, error: str):
+        integration_str = integration.provider + (
+            "." + integration.method if integration.method else ""
+        )
+        super().__init__(
+            f"Error in executing {integration_str}: {error}",
+            http_code=500,
+        )