pygeai 0.4.0b11__tar.gz → 0.5.0b3__tar.gz

{pygeai-0.4.0b11/pygeai.egg-info → pygeai-0.5.0b3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pygeai
-Version: 0.4.0b11
+Version: 0.5.0b3
 Summary: Software Development Kit to interact with Globant Enterprise AI.
 Author-email: Globant <geai-sdk@globant.com>
 License-Expression: MIT
@@ -48,7 +48,7 @@ By using the Python SDK to interact with Globant Enterprise AI, you agree with t
 [Terms and Conditions](https://www.globant.com/enterprise-ai/terms-of-use)
 
 ## Compatibility
-This package is compatible with the Globant Enterprise AI release from June 2025.
+This package is compatible with the Globant Enterprise AI release from November 2025.
 
 ## Configuration
 
@@ -60,22 +60,50 @@ Before using the SDK, you need to define `GEAI_API_KEY` (`$SAIA_APITOKEN`) and `
 
 **Note:** If you plan to use the [Evaluation Module](https://wiki.genexus.com/enterprise-ai/wiki?896,Evaluation), you must also define `GEAI_API_EVAL_URL`
 
+### Credentials file
+The credentials file is organized via profiles, so one can interact with different instances of Globant Enterprise AI by just
+referencing which profile one wishes to use. This also applies to different levels of permissions.
+
+The structure of the credentials file is as follows:
+
+```
+[default]
+geai_api_key = <API_TOKEN>
+geai_api_base_url = <GEAI_BASE_URL>
+geai_api_evaul_url = <GEAI_EVALUATION_MODULE_URL>
+
+[<another alias>]
+geai_api_key = <API_TOKEN>
+geai_api_base_url = <GEAI_BASE_URL>
+geai_api_evaul_url = <GEAI_EVALUATION_MODULE_URL>
+
+```
+
+After setting the profiles, one can use them with the --alias option, for example:
+
+```bash 
+geai --alias test llm list-providers
+```
+
 ## Modules
 
-The SDK consists of several modules, all accessible through a meta-package:
-
-- **`pygeai`**: This meta-package encapsulates all components of the SDK.
-- **`pygeai-cli`**: This package provides a command-line tool for interacting with the SDK.
-- **`pygeai-chat`**: This package offers facilities to chat with assistants/agents created in Globant Enterprise AI.
-- **`pygeai-dbg`**: This package includes a debugger to troubleshoot potential SDK issues and gain detailed insights into its operations.
-- **`pygeai-core`**: This package handles interactions with the fundamental components of Globant Enterprise AI, including users, groups, permissions, API keys, organizations, and [Projects](https://wiki.genexus.com/enterprise-ai/wiki?565,Projects).
-- **`pygeai-admin`**: This package enables interactions with the Globant Enterprise AI instance.
-- **`pygeai-lab`**: This package facilitates interactions with AI LAB.
-- **`pygeai-evaluation`**: This package provides functionality from the evaluation module.
-- **`pygeai-gam`**: This package allows interaction with [GAM] (https://wiki.genexus.com/commwiki/wiki?24746,Table+of+contents%3AGeneXus+Access+Manager+%28GAM%29,).
-- **`pygeai-assistant`**: This package handles interactions with various Assistants, including [Data Analyst Assistants](https://wiki.genexus.com/enterprise-ai/wiki?886,Data+Analyst+Assistant+2.0), [RAG Assistants](https://wiki.genexus.com/enterprise-ai/wiki?44,RAG+Assistants+Introduction), [Chat with Data Assistants](https://wiki.genexus.com/enterprise-ai/wiki?159,Chat+with+Data+Assistant), [Chat with API Assistants](https://wiki.genexus.com/enterprise-ai/wiki?110,API+Assistant), and [Chat Assistants](https://wiki.genexus.com/enterprise-ai/wiki?708,Chat+Assistant).
-- **`pygeai-organization`**: This package facilitates interactions with Organizations in Globant Enterprise AI.
-- **`pygeai-flows`**: This package enables interactions with [Flows](https://wiki.genexus.com/enterprise-ai/wiki?321,Flows+in+Globant+Enterprise+AI) [in development]. 
+The SDK consists of several modules, all accessible through a single package `pygeai`:
+
+- **`pygeai`**: The package encapsulates all components of the SDK.
+- **`admin`**: Allows the user to interact with some of the administration endpoints of Globant Enterprise AI.
+- **`assistant`**: This module handles interactions with various Assistants, including [Data Analyst Assistants](https://wiki.genexus.com/enterprise-ai/wiki?886,Data+Analyst+Assistant+2.0), [RAG Assistants](https://wiki.genexus.com/enterprise-ai/wiki?44,RAG+Assistants+Introduction), [Chat with Data Assistants](https://wiki.genexus.com/enterprise-ai/wiki?159,Chat+with+Data+Assistant), [Chat with API Assistants](https://wiki.genexus.com/enterprise-ai/wiki?110,API+Assistant), and [Chat Assistants](https://wiki.genexus.com/enterprise-ai/wiki?708,Chat+Assistant [soon to be deprecated]).
+- **`chat`**: This module offers facilities to chat with assistants/agents created in Globant Enterprise AI.
+- **`core`**: This module handles interactions with the fundamental components of Globant Enterprise.
+- **`cli`**: This module provides a command-line tool for interacting with the SDK.
+- **`dbg`**: This module includes a debugger to troubleshoot potential SDK issues and gain detailed insights into its operations.
+- **`evaluation`**: This module provides functionality for the evaluation module.
+- **`flows`**: This module enables interactions with [Flows](https://wiki.genexus.com/enterprise-ai/wiki?321,Flows+in+Globant+Enterprise+AI) [in development]. 
+- **`gam`**: This module allows interaction with [GAM] (https://wiki.genexus.com/commwiki/wiki?24746,Table+of+contents%3AGeneXus+Access+Manager+%28GAM%29,).
+- **`health`**: Provides an easy way to check the status of the Globant Enterprise AI instance one is using.
+- **`lab`**: This module facilitates interactions with [The Lab] (https://docs.globant.ai/en/wiki?1671,The+Lab).
+- **`organization`**: This module facilitates interactions with Organizations and [Projects](https://wiki.genexus.com/enterprise-ai/wiki?565,Projects) in Globant Enterprise AI.
+- **`migration`**: This module provides functionality to migrate Agents, Tools and Projects between instances and organizations.
+- **`proxy`**: This module handles [integrations with MCP and A2A] (https://docs.globant.ai/en/wiki?1179,Importing+Tools+using+MCP+and+A2A+Servers).
 
 ## Usage
 

{pygeai-0.4.0b11 → pygeai-0.5.0b3}/README.md

@@ -9,7 +9,7 @@ By using the Python SDK to interact with Globant Enterprise AI, you agree with t
 [Terms and Conditions](https://www.globant.com/enterprise-ai/terms-of-use)
 
 ## Compatibility
-This package is compatible with the Globant Enterprise AI release from June 2025.
+This package is compatible with the Globant Enterprise AI release from November 2025.
 
 ## Configuration
 
@@ -21,22 +21,50 @@ Before using the SDK, you need to define `GEAI_API_KEY` (`$SAIA_APITOKEN`) and `
 
 **Note:** If you plan to use the [Evaluation Module](https://wiki.genexus.com/enterprise-ai/wiki?896,Evaluation), you must also define `GEAI_API_EVAL_URL`
 
+### Credentials file
+The credentials file is organized via profiles, so one can interact with different instances of Globant Enterprise AI by just
+referencing which profile one wishes to use. This also applies to different levels of permissions.
+
+The structure of the credentials file is as follows:
+
+```
+[default]
+geai_api_key = <API_TOKEN>
+geai_api_base_url = <GEAI_BASE_URL>
+geai_api_evaul_url = <GEAI_EVALUATION_MODULE_URL>
+
+[<another alias>]
+geai_api_key = <API_TOKEN>
+geai_api_base_url = <GEAI_BASE_URL>
+geai_api_evaul_url = <GEAI_EVALUATION_MODULE_URL>
+
+```
+
+After setting the profiles, one can use them with the --alias option, for example:
+
+```bash 
+geai --alias test llm list-providers
+```
+
 ## Modules
 
-The SDK consists of several modules, all accessible through a meta-package:
-
-- **`pygeai`**: This meta-package encapsulates all components of the SDK.
-- **`pygeai-cli`**: This package provides a command-line tool for interacting with the SDK.
-- **`pygeai-chat`**: This package offers facilities to chat with assistants/agents created in Globant Enterprise AI.
-- **`pygeai-dbg`**: This package includes a debugger to troubleshoot potential SDK issues and gain detailed insights into its operations.
-- **`pygeai-core`**: This package handles interactions with the fundamental components of Globant Enterprise AI, including users, groups, permissions, API keys, organizations, and [Projects](https://wiki.genexus.com/enterprise-ai/wiki?565,Projects).
-- **`pygeai-admin`**: This package enables interactions with the Globant Enterprise AI instance.
-- **`pygeai-lab`**: This package facilitates interactions with AI LAB.
-- **`pygeai-evaluation`**: This package provides functionality from the evaluation module.
-- **`pygeai-gam`**: This package allows interaction with [GAM] (https://wiki.genexus.com/commwiki/wiki?24746,Table+of+contents%3AGeneXus+Access+Manager+%28GAM%29,).
-- **`pygeai-assistant`**: This package handles interactions with various Assistants, including [Data Analyst Assistants](https://wiki.genexus.com/enterprise-ai/wiki?886,Data+Analyst+Assistant+2.0), [RAG Assistants](https://wiki.genexus.com/enterprise-ai/wiki?44,RAG+Assistants+Introduction), [Chat with Data Assistants](https://wiki.genexus.com/enterprise-ai/wiki?159,Chat+with+Data+Assistant), [Chat with API Assistants](https://wiki.genexus.com/enterprise-ai/wiki?110,API+Assistant), and [Chat Assistants](https://wiki.genexus.com/enterprise-ai/wiki?708,Chat+Assistant).
-- **`pygeai-organization`**: This package facilitates interactions with Organizations in Globant Enterprise AI.
-- **`pygeai-flows`**: This package enables interactions with [Flows](https://wiki.genexus.com/enterprise-ai/wiki?321,Flows+in+Globant+Enterprise+AI) [in development]. 
+The SDK consists of several modules, all accessible through a single package `pygeai`:
+
+- **`pygeai`**: The package encapsulates all components of the SDK.
+- **`admin`**: Allows the user to interact with some of the administration endpoints of Globant Enterprise AI.
+- **`assistant`**: This module handles interactions with various Assistants, including [Data Analyst Assistants](https://wiki.genexus.com/enterprise-ai/wiki?886,Data+Analyst+Assistant+2.0), [RAG Assistants](https://wiki.genexus.com/enterprise-ai/wiki?44,RAG+Assistants+Introduction), [Chat with Data Assistants](https://wiki.genexus.com/enterprise-ai/wiki?159,Chat+with+Data+Assistant), [Chat with API Assistants](https://wiki.genexus.com/enterprise-ai/wiki?110,API+Assistant), and [Chat Assistants](https://wiki.genexus.com/enterprise-ai/wiki?708,Chat+Assistant [soon to be deprecated]).
+- **`chat`**: This module offers facilities to chat with assistants/agents created in Globant Enterprise AI.
+- **`core`**: This module handles interactions with the fundamental components of Globant Enterprise.
+- **`cli`**: This module provides a command-line tool for interacting with the SDK.
+- **`dbg`**: This module includes a debugger to troubleshoot potential SDK issues and gain detailed insights into its operations.
+- **`evaluation`**: This module provides functionality for the evaluation module.
+- **`flows`**: This module enables interactions with [Flows](https://wiki.genexus.com/enterprise-ai/wiki?321,Flows+in+Globant+Enterprise+AI) [in development]. 
+- **`gam`**: This module allows interaction with [GAM] (https://wiki.genexus.com/commwiki/wiki?24746,Table+of+contents%3AGeneXus+Access+Manager+%28GAM%29,).
+- **`health`**: Provides an easy way to check the status of the Globant Enterprise AI instance one is using.
+- **`lab`**: This module facilitates interactions with [The Lab] (https://docs.globant.ai/en/wiki?1671,The+Lab).
+- **`organization`**: This module facilitates interactions with Organizations and [Projects](https://wiki.genexus.com/enterprise-ai/wiki?565,Projects) in Globant Enterprise AI.
+- **`migration`**: This module provides functionality to migrate Agents, Tools and Projects between instances and organizations.
+- **`proxy`**: This module handles [integrations with MCP and A2A] (https://docs.globant.ai/en/wiki?1179,Importing+Tools+using+MCP+and+A2A+Servers).
 
 ## Usage
 

{pygeai-0.4.0b11 → pygeai-0.5.0b3}/pygeai/__init__.py

@@ -4,7 +4,7 @@ import sys
 from pathlib import Path
 
 __author__ = 'Globant'
-__version__ = '0.4.0'
+__version__ = '0.5.0'
 
 # Add vendor directory to Python path
 package_root = Path(__file__).parent

{pygeai-0.4.0b11 → pygeai-0.5.0b3}/pygeai/cli/__init__.py

@@ -4,5 +4,5 @@ GEAI - ClI
 Command line interface to interact with Globant Enterprise AI.
 """
 
-__version__ = '0.4.0'
+__version__ = '0.5.0'
 

{pygeai-0.4.0b11 → pygeai-0.5.0b3}/pygeai/core/base/session.py

@@ -30,7 +30,7 @@ class Session(metaclass=Singleton):
             eval_url: Optional[str] = None,
     ):
         if not api_key or not base_url:
-            raise MissingRequirementException("Cannot instantiate session without api_key and base_url")
+            logger.warning("Cannot instantiate session without api_key and base_url")
 
         self.__api_key = api_key
         self.__base_url = base_url

{pygeai-0.4.0b11 → pygeai-0.5.0b3}/pygeai/core/common/config.py

@@ -114,9 +114,7 @@ def get_settings():
 
 if __name__ == "__main__":
     settings = get_settings()
-    geai_api_key = settings.get_api_key()
     geai_base_url = settings.get_base_url()
     geai_eval_url = settings.get_eval_url()
-    print(f"api_key: {geai_api_key}")
     print(f"base_url: {geai_base_url}")
     print(f"eval_url: {geai_eval_url}")

{pygeai-0.4.0b11 → pygeai-0.5.0b3}/pygeai/lab/managers.py

@@ -559,7 +559,7 @@ class AILabManager:
             tool_id: Optional[str] = None,
             tool_public_name: Optional[str] = None,
             parameters: List[ToolParameter] = None
-    ) -> Tool:
+    ) -> EmptyResponse:
         """
         Sets or updates parameters for a specific tool in the specified project.
 
@@ -591,7 +591,7 @@ class AILabManager:
             logger.error(f"Error received while setting tool parameters: {error}")
             raise APIError(f"Error received while setting tool parameters: {error}")
 
-        result = ToolMapper.map_to_tool(response_data)
+        result = ResponseMapper.map_to_empty_response(response_data or "Parameter set successfully")
         return result
 
     def list_reasoning_strategies(

{pygeai-0.4.0b11 → pygeai-0.5.0b3}/pygeai/lab/tools/clients.py

@@ -114,7 +114,8 @@ class ToolClient(AILabClient):
             headers=headers,
             data=data
         )
-        if response.status_code != 200:
+        
+        if response.status_code >= 300:
             logger.error(f"Invalid status code returned from the API endpoint: {response.text}")
             raise APIResponseError(f"API returned an error: {response.text}")
         try:
@@ -168,7 +169,7 @@ class ToolClient(AILabClient):
                 "allowExternal": allow_external
             }
         )
-        if response.status_code != 200:
+        if response.status_code >= 300:
             logger.error(f"Invalid status code returned from the API endpoint: {response.text}")
             raise APIResponseError(f"API returned an error: {response.text}")
         try:
@@ -212,7 +213,7 @@ class ToolClient(AILabClient):
                 "allowDrafts": allow_drafts,
             }
         )
-        if response.status_code != 200:
+        if response.status_code >= 300:
             logger.error(f"Invalid status code returned from the API endpoint: {response.text}")
             raise APIResponseError(f"API returned an error: {response.text}")
         try:
@@ -255,7 +256,7 @@ class ToolClient(AILabClient):
             endpoint=endpoint,
             headers=headers
         )
-        if response.status_code != 200:
+        if response.status_code >= 300:
             logger.error(f"Invalid status code returned from the API endpoint: {response.text}")
             raise APIResponseError(f"API returned an error: {response.text}")
 
@@ -375,7 +376,7 @@ class ToolClient(AILabClient):
             headers=headers,
             data=data
         )
-        if response.status_code != 200:
+        if response.status_code >= 300:
             logger.error(f"Invalid status code returned from the API endpoint: {response.text}")
             raise APIResponseError(f"API returned an error: {response.text}")
         try:
@@ -463,7 +464,7 @@ class ToolClient(AILabClient):
                 "allowDrafts": allow_drafts,
             }
         )
-        if response.status_code != 200:
+        if response.status_code >= 300:
             logger.error(f"Invalid status code returned from the API endpoint: {response.text}")
             raise APIResponseError(f"API returned an error: {response.text}")
         try:
@@ -520,16 +521,11 @@ class ToolClient(AILabClient):
             headers=headers,
             data=data
         )
-        if response.status_code != 200:
-            logger.error(f"Invalid status code returned from the API endpoint: {response.text}")
-            raise APIResponseError(f"API returned an error: {response.text}")
-        try:
-            result = response.json()
-        except JSONDecodeError as e:
-            logger.error(f"Unable to set parameters for tool {tool_id or tool_public_name} in project {self.project_id}: JSON parsing error (status {response.status_code}): {e}. Response: {response.text}")
-            raise InvalidAPIResponseException(f"Unable to set parameters for tool {tool_id or tool_public_name} in project {self.project_id}: {response.text}")
-
-        return result
+        if response.status_code != 204:
+            logger.error(f"Unable to set parameter for tool {tool_id or tool_public_name} from project {self.project_id}: JSON parsing error (status {response.status_code}). Response: {response.text}")
+            raise InvalidAPIResponseException(f"Unable to set parameter for tool {tool_id or tool_public_name} from project {self.project_id}: {response.text}")
+        else:
+            return {}
 
     def export_tool(
             self,
@@ -559,7 +555,7 @@ class ToolClient(AILabClient):
             endpoint=endpoint,
             headers=headers,
         )
-        if response.status_code != 200:
+        if response.status_code >= 300:
             logger.error(f"Invalid status code returned from the API endpoint: {response.text}")
             raise APIResponseError(f"API returned an error: {response.text}")
         try:

pygeai-0.5.0b3/pygeai/tests/integration/assistants/rag/test_create_rag.py

@@ -0,0 +1,91 @@
+from unittest import TestCase
+import uuid
+from pygeai.assistant.managers import AssistantManager
+from pygeai.core.models import WelcomeData, LlmSettings
+from pygeai.assistant.rag.models import (
+    Search,
+    RetrieverOptions,
+    SearchOptions,
+    ChunkOptions,
+    IndexOptions,
+    RAGAssistant,
+)
+
+
+class TestAssistantCreateRagIntegration(TestCase):
+
+    def setUp(self):
+        """
+        Set up the test environment.
+        """
+        self.assistant_manager = AssistantManager(alias="beta")
+
+        self.new_rag = self.__load_rag()
+        self.created_rag: RAGAssistant = None
+
+
+    def tearDown(self):
+        """
+        Clean up after each test if necessary.
+        This can be used to delete the created tool
+        """
+        if isinstance(self.created_rag, RAGAssistant):
+            self.assistant_manager.delete_assistant(assistant_name=self.created_rag.name)            
+
+
+    def __load_rag(self):
+        llm_options = LlmSettings(
+            cache=False,
+            temperature=0.1,
+            max_tokens=999,
+            model_name="gpt-3.5-turbo-16k",
+            n=1,
+            presence_penalty=0,
+            frequency_penalty=0,
+            provider="OpenAI",
+            stream=False,
+            top_p=1.0,
+            type=None,
+            verbose=True,
+        )
+
+        retriever_options = RetrieverOptions(type="vectorStore")
+
+        search_options = SearchOptions(
+            history_count=2,
+            llm=llm_options,
+            search=Search(
+                k=5,
+                return_source_documents=False,
+                score_threshold=0,
+                prompt="Use {context} and {question}",
+                template="",
+            ),
+            retriever=retriever_options,
+        )
+
+        chunk_options = ChunkOptions(chunk_size=999, chunk_overlap=0)
+
+        index_options = IndexOptions(chunks=chunk_options)
+
+        welcome_data = WelcomeData(
+            title="Test Profile Welcome Data",
+            description="Test Profile with WelcomeData",
+            features=[],
+            examples_prompt=[],
+        )
+
+        return RAGAssistant(
+            name=str(uuid.uuid4()),
+            description="Test Profile with WelcomeData",
+            search_options=search_options,
+            index_options=index_options,
+            welcome_data=welcome_data,
+        )
+
+
+    def test_create_rag_assistant(self):
+        rag_assistant = self.__load_rag()
+        self.created_rag = self.assistant_manager.create_assistant(rag_assistant)
+
+        self.assertIsInstance(self.created_rag, RAGAssistant, "Failed to create RAG assistant")

{pygeai-0.4.0b11 → pygeai-0.5.0b3}/pygeai/tests/integration/chat/test_generate_image.py

@@ -56,13 +56,9 @@ class TestChatGenerateImageIntegration(TestCase):
         self.new_image["model"] = ""
         created_image = self.__generate_image()
 
-        self.assertEqual(
-            created_image["error"]["code"], 400,
-            "Expected a 400 code for no model"
-        )
         self.assertEqual(
             created_image["error"]["message"],
-            "Invalid 'model' name. Must follow pattern {provider}/{modelName}", 
+            "LLM Provider NOT provided. Pass in the LLM provider you are trying to call", 
             "Expected an error message when no model is provided"
         )
 

{pygeai-0.4.0b11 → pygeai-0.5.0b3}/pygeai/tests/integration/lab/agents/test_create_agent.py

@@ -33,7 +33,6 @@ class TestAILabCreateAgentIntegration(TestCase):
             access_scope="public",
             public_name=f"public_{random_str}",
             job_description="Translator",
-            avatar_image="https://www.shareicon.net/data/128x128/2016/11/09/851442_logo_512x512.png",
             description="Agent that translates from any language to english.",
             agent_data=AgentData(
                 prompt=Prompt(
@@ -142,7 +141,6 @@ class TestAILabCreateAgentIntegration(TestCase):
         self.assertEqual(prompt.inputs, agent.agent_data.prompt.inputs)
 
 
-    @unittest.skip("Skipping test due to mismatch in expected behavior")
     def test_create_agent_without_required_data(self):
         test_params = [ True, False ]
 
@@ -152,12 +150,19 @@ class TestAILabCreateAgentIntegration(TestCase):
                 self.new_agent = Agent(
                     name=str(uuid.uuid4())
                 )
-                with self.assertRaises(APIError) as exception:
-                    self.__create_agent(automatic_publish=auto_publish)
-
-                #TODO: Change validation error to a more specific one
-                self.assertIn("400",str(exception.exception))    
-                self.assertIn("Bad Request",str(exception.exception))
+                if auto_publish:
+                    
+                    with self.assertRaises(APIError) as exception:
+                        self.__create_agent(automatic_publish=auto_publish)
+
+                    #TODO: Change validation error to a more specific one
+                    self.assertIn(
+                        "A valid prompt is required. To be valid, it must provide clear instructions to the model",
+                        str(exception.exception)
+                    )   
+                else:
+                    created_agent = self.__create_agent(automatic_publish=auto_publish)
+                    self.assertTrue(isinstance(created_agent, Agent), "Expected a created agent")
 
 
     def test_create_agent_no_name(self):

{pygeai-0.4.0b11 → pygeai-0.5.0b3}/pygeai/tests/integration/lab/agents/test_create_sharing_link.py

@@ -1,4 +1,5 @@
 from unittest import TestCase
+import unittest
 from pygeai.lab.managers import AILabManager
 from pygeai.lab.models import SharingLink
 from pygeai.core.common.exceptions import APIError, MissingRequirementException
@@ -16,7 +17,7 @@ class TestAILabCreateSharingLinkIntegration(TestCase):
             agent_id=self.agent_id if agent_id is None else agent_id
         )
     
-
+    @unittest.skip("Endpoint is not found. Validate if it was deleted")
     def test_create_sharing_link(self):    
         shared_link = self.__create_sharing_link()        
         self.assertIsInstance(shared_link, SharingLink, "Expected response to be an instance of SharingLink")
@@ -46,6 +47,7 @@ class TestAILabCreateSharingLinkIntegration(TestCase):
         )
 
 
+    @unittest.skip("Endpoint is not found. Validate if it was deleted")
     def test_create_sharing_link_no_agent_id(self):
         with self.assertRaises(MissingRequirementException) as context:
             self.__create_sharing_link(agent_id="")
@@ -56,6 +58,7 @@ class TestAILabCreateSharingLinkIntegration(TestCase):
         )
 
 
+    @unittest.skip("Endpoint is not found. Validate if it was deleted")
     def test_create_sharing_link_invalid_agent_id(self):
         with self.assertRaises(APIError) as exception:
             invalid_id = "0026e53d-ea78-4cac-af9f-12650invalid"

{pygeai-0.4.0b11 → pygeai-0.5.0b3}/pygeai/tests/integration/lab/agents/test_publish_agent_revision.py

@@ -28,7 +28,6 @@ class TestAILabPublishAgentRevisionIntegration(TestCase):
             access_scope="public",
             public_name=f"public_{random_str}",
             job_description=f"SummarizerAgent{random_str}",
-            avatar_image="https://www.shareicon.net/data/128x128/2016/11/09/851442_logo_512x512.png",
             description=f"Agent that summarized documents. {random_str}",
             agent_data=AgentData(
                 prompt=Prompt(

{pygeai-0.4.0b11 → pygeai-0.5.0b3}/pygeai/tests/integration/lab/agents/test_update_agent.py

@@ -25,7 +25,6 @@ class TestAILabUpdateAgentIntegration(TestCase):
             access_scope="public",
             public_name=f"public_{random_str}",
             job_description=f"SummarizerAgent{random_str}",
-            avatar_image="https://www.shareicon.net/data/128x128/2016/11/09/851442_logo_512x512.png",
             description=f"Agent that summarized documents. {random_str}",
             agent_data=AgentData(
                 prompt=Prompt(
@@ -192,23 +191,21 @@ class TestAILabUpdateAgentIntegration(TestCase):
         self.agent_to_update.agent_data.prompt.instructions = ""
 
         for auto_publish in test_params:
-            with self.subTest(input=auto_publish): 
-                with self.assertRaises(ValidationError) as exception:
-                    self.__update_agent(automatic_publish=auto_publish)
-                
-                self.assertIn(
-                    "instructions",
-                    str(exception.exception),
-                    f"Expected a validation error about allowed values for instructions when autopublish is {'enabled' if auto_publish else 'disabled'}"
-                )
-                self.assertIn(
-                    "Input should be a valid string [type=string_type, input_value=None, input_type=NoneType]",
-                    str(exception.exception),
-                    f"Expected a validation error about allowed values for instructions when autopublish is {'enabled' if auto_publish else 'disabled'}"
-                )
+            with self.subTest(input=auto_publish):
+                if auto_publish == True:
+                    with self.assertRaises(ValidationError) as exception:
+                        self.__update_agent(automatic_publish=auto_publish)
+                    self.assertIn(
+                        "agent_data.prompt must have at least instructions for publication",
+                        str(exception.exception),
+                        f"Expected a validation error about allowed values for instructions when autopublish is {'enabled' if auto_publish else 'disabled'}"
+                    )
+                else:
+                    updated_agent = self.__update_agent(automatic_publish=auto_publish)
+                    self.assertTrue(isinstance(updated_agent, Agent))
 
-    # TODO: Change validation when API behavior is fixed
-    def test_update_agent_no_model(self):        
+
+    def test_update_agent_no_model(self):
         test_params = [ True, False ]
         self.agent_to_update.agent_data.models[0].name = ""
         
@@ -218,9 +215,8 @@ class TestAILabUpdateAgentIntegration(TestCase):
                 # If the agent is not published, the API returns a warning message for invalid model name. However, the sdk mapping is not returning it.
                 if auto_publish == False:                    
                     updated_agent = self.__update_agent(automatic_publish=auto_publish)
-                    error_msg = str(exception.exception)
                     
-                    self.assertTrue(self.isInstance(updated_agent), Agent)
+                    self.assertTrue(isinstance(updated_agent, Agent))
                 else:                 
                     with self.assertRaises(APIError) as exception:
                         self.__update_agent(automatic_publish=auto_publish)