versionhq 1.2.3.4__tar.gz → 1.2.3.6__tar.gz

{versionhq-1.2.3.4 → versionhq-1.2.3.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: versionhq
-Version: 1.2.3.4
+Version: 1.2.3.6
 Summary: An agentic orchestration framework for building agent networks that handle task automation.
 Author-email: Kuriko Iwai <kuriko@versi0n.io>
 License: MIT License
@@ -68,6 +68,9 @@ Requires-Dist: composio-core==0.7.0
 Requires-Dist: networkx>=3.4.2
 Requires-Dist: matplotlib>=3.10.0
 Requires-Dist: boto3>=1.37.1
+Requires-Dist: scikit-learn>=1.6.1
+Requires-Dist: numpy>=1.26.4
+Requires-Dist: pandas>=2.2.3
 Provides-Extra: docling
 Requires-Dist: docling>=2.25.2; extra == "docling"
 Provides-Extra: mem0ai
@@ -82,10 +85,6 @@ Requires-Dist: sec-api>=1.0.28; extra == "tools"
 Provides-Extra: torch
 Requires-Dist: torch>=2.6.0; extra == "torch"
 Requires-Dist: torchvision>=0.21.0; extra == "torch"
-Provides-Extra: evals
-Requires-Dist: scikit-learn>=1.6.1; extra == "evals"
-Requires-Dist: numpy>=1.26.4; extra == "evals"
-Requires-Dist: pandas>=2.2.3; extra == "evals"
 
 # Overview
 
@@ -143,9 +142,9 @@ Agentic orchestration framework for multi-agent networks and task graphs for com
 
 ## Key Features
 
-`versionhq` is a Python framework for agent networks that handle complex task automation without human interaction.
+`versionhq` is a Python framework designed for automating complex, multi-step tasks using autonomous agent networks.
 
-Agents are model-agnostic, and will improve task output, while optimizing token cost and job latency, by sharing their memory, knowledge base, and RAG tools with other agents in the network.
+Users can either configure their agents and network manually or allow the system to automatically manage the process based on provided task goals.
 
 
 ###  Agent Network
@@ -172,7 +171,7 @@ Each node is triggered by specific events and executed by an assigned agent once
 While the network automatically reconfigures itself, you retain the ability to direct the agents using `should_reform` variable.
 
 
-The following code snippet demonstrates the `TaskGraph` and its visualization, saving the diagram to the `uploads` directory.
+The following code snippet explicitly demonstrates the `TaskGraph` and its visualization, saving the diagram to the `uploads` directory.
 
 ```python
 import versionhq as vhq
@@ -214,15 +213,13 @@ A `TaskGraph` represents tasks as `nodes` and their execution dependencies as `e
 
 `Agent Networks` can handle `TaskGraph` objects by optimizing their formations.
 
-The following example demonstrates a simple concept of a `supervising` agent network handling a task graph with three tasks and one critical edge.
-
 <img src="https://res.cloudinary.com/dfeirxlea/image/upload/v1739337639/pj_m_home/zfg4ccw1m1ww1tpnb0pa.png">
 
 <hr />
 
 ### Optimization
 
-Agents are model-agnostic and can handle multiple tasks, leveraging their own and their peers' knowledge sources, memories, and tools.
+Autonomous agents are model-agnostic and can leverage their own and their peers' knowledge sources, memories, and tools.
 
 Agents are optimized during network formation, but customization is possible before or after.
 
@@ -581,7 +578,9 @@ Common issues and solutions:
 
 * Issues related to dependencies: `rm -rf uv.lock`, `uv cache clean`, `uv venv`, and run `uv pip install -r requirements.txt -v`.
 
-* Issues related to agents and other systems: Check `.logs` directory located in the root directory for error messages and stack traces.
+* Issues related to `torch` installation: Add optional dependencies by `uv add versionhq[torch]`.
+
+* Issues related to agents and other systems: Check `.logs` directory located at the root of the project directory for error messages and stack traces.
 
 * Issues related to `Python quit unexpectedly`: Check [this stackoverflow article](https://stackoverflow.com/questions/59888499/macos-catalina-python-quit-unexpectedly-error).
 

{versionhq-1.2.3.4 → versionhq-1.2.3.6}/README.md

@@ -54,9 +54,9 @@ Agentic orchestration framework for multi-agent networks and task graphs for com
 
 ## Key Features
 
-`versionhq` is a Python framework for agent networks that handle complex task automation without human interaction.
+`versionhq` is a Python framework designed for automating complex, multi-step tasks using autonomous agent networks.
 
-Agents are model-agnostic, and will improve task output, while optimizing token cost and job latency, by sharing their memory, knowledge base, and RAG tools with other agents in the network.
+Users can either configure their agents and network manually or allow the system to automatically manage the process based on provided task goals.
 
 
 ###  Agent Network
@@ -83,7 +83,7 @@ Each node is triggered by specific events and executed by an assigned agent once
 While the network automatically reconfigures itself, you retain the ability to direct the agents using `should_reform` variable.
 
 
-The following code snippet demonstrates the `TaskGraph` and its visualization, saving the diagram to the `uploads` directory.
+The following code snippet explicitly demonstrates the `TaskGraph` and its visualization, saving the diagram to the `uploads` directory.
 
 ```python
 import versionhq as vhq
@@ -125,15 +125,13 @@ A `TaskGraph` represents tasks as `nodes` and their execution dependencies as `e
 
 `Agent Networks` can handle `TaskGraph` objects by optimizing their formations.
 
-The following example demonstrates a simple concept of a `supervising` agent network handling a task graph with three tasks and one critical edge.
-
 <img src="https://res.cloudinary.com/dfeirxlea/image/upload/v1739337639/pj_m_home/zfg4ccw1m1ww1tpnb0pa.png">
 
 <hr />
 
 ### Optimization
 
-Agents are model-agnostic and can handle multiple tasks, leveraging their own and their peers' knowledge sources, memories, and tools.
+Autonomous agents are model-agnostic and can leverage their own and their peers' knowledge sources, memories, and tools.
 
 Agents are optimized during network formation, but customization is possible before or after.
 
@@ -492,7 +490,9 @@ Common issues and solutions:
 
 * Issues related to dependencies: `rm -rf uv.lock`, `uv cache clean`, `uv venv`, and run `uv pip install -r requirements.txt -v`.
 
-* Issues related to agents and other systems: Check `.logs` directory located in the root directory for error messages and stack traces.
+* Issues related to `torch` installation: Add optional dependencies by `uv add versionhq[torch]`.
+
+* Issues related to agents and other systems: Check `.logs` directory located at the root of the project directory for error messages and stack traces.
 
 * Issues related to `Python quit unexpectedly`: Check [this stackoverflow article](https://stackoverflow.com/questions/59888499/macos-catalina-python-quit-unexpectedly-error).
 

{versionhq-1.2.3.4 → versionhq-1.2.3.6}/docs/core/agent/task-handling.md

@@ -25,7 +25,7 @@ agent = vhq.Agent(
 	goal="Coping with price competition in saturated markets"
 )
 
-assert agent.backstory == "You are an expert marketing analyst with relevant skillsets and abilities to query relevant information from the given knowledge sources. Leveraging these, you will identify competitive solutions to achieve the following goal: coping with price competition in saturated markets."
+assert agent.backstory == "You are an expert marketing analyst with relevant skills and abilities to query relevant information from the given knowledge sources. Leveraging these, you will identify competitive solutions to achieve the following goal: coping with price competition in saturated markets."
 ```
 
 You can also specify your own backstory by simply adding the value to the backstory field of the Agent model:

{versionhq-1.2.3.4 → versionhq-1.2.3.6}/docs/index.md

@@ -23,9 +23,9 @@ A Python framework for agentic orchestration that handles complex task automatio
 
 ## Key Features
 
-`versionhq` is a Python framework for agent networks that handle complex task automation without human interaction.
+`versionhq` is a Python framework designed for automating complex, multi-step tasks using autonomous agent networks.
 
-Agents are model-agnostic, and will improve task output, while optimizing token cost and job latency, by sharing their memory, knowledge base, and RAG tools with other agents in the network.
+Users can either configure their agents and network manually or allow the system to automatically manage the process based on provided task goals.
 
 
 ### Agent Network
@@ -51,8 +51,7 @@ Each node is triggered by specific events and executed by an assigned agent once
 
 While the network automatically reconfigures itself, you retain the ability to direct the agents using `should_reform` variable.
 
-
-The following code snippet demonstrates the `TaskGraph` and its visualization, saving the diagram to the `uploads` directory.
+The following code snippet explicitly demonstrates the `TaskGraph` and its visualization, saving the diagram to the `uploads` directory.
 
 ```python
 import versionhq as vhq
@@ -94,8 +93,6 @@ A `TaskGraph` represents tasks as `nodes` and their execution dependencies as `e
 
 `Agent Networks` can handle `TaskGraph` objects by optimizing their formations.
 
-The following example demonstrates a simple concept of a `supervising` agent network handling a task graph with three tasks and one critical edge.
-
 <img src="https://res.cloudinary.com/dfeirxlea/image/upload/v1739337639/pj_m_home/zfg4ccw1m1ww1tpnb0pa.png">
 
 * Ref: <a href="/core/task-graph">TaskGraph</a> class
@@ -104,7 +101,7 @@ The following example demonstrates a simple concept of a `supervising` agent net
 
 ### Optimization
 
-Agents are model-agnostic and can handle multiple tasks, leveraging their own and their peers' knowledge sources, memories, and tools.
+Autonomous agents are model-agnostic and can leverage their own and their peers' knowledge sources, memories, and tools.
 
 Agents are optimized during network formation, but customization is possible before or after.
 
@@ -239,12 +236,15 @@ Common issues and solutions:
 
 * Issues related to dependencies: `rm -rf uv.lock`, `uv cache clean`, `uv venv`, and run `uv pip install -r requirements.txt -v`.
 
-* Issues related to agents and other systems: Check `.logs` directory located in the root directory for detailed error messages and stack traces.
+* Issues related to `torch` installation: Add optional dependencies by `uv add versionhq[torch]`.
+
+* Issues related to agents and other systems: Check `.logs` directory located at the root of the project directory for error messages and stack traces.
 
 * Issues related to `Python quit unexpectedly`: Check [this stackoverflow article](https://stackoverflow.com/questions/59888499/macos-catalina-python-quit-unexpectedly-error).
 
 * `reportMissingImports` error from pyright after installing the package: This might occur when installing new libraries while VSCode is running. Open the command pallete (ctrl + shift + p) and run the Python: Restart language server task.
 
+
 <hr />
 
 ## FAQ

{versionhq-1.2.3.4 → versionhq-1.2.3.6}/pyproject.toml

@@ -15,7 +15,7 @@ exclude = ["test*", "__pycache__", "*.egg-info"]
 
 [project]
 name = "versionhq"
-version = "1.2.3.4"
+version = "1.2.3.6"
 authors = [{ name = "Kuriko Iwai", email = "kuriko@versi0n.io" }]
 description = "An agentic orchestration framework for building agent networks that handle task automation."
 readme = "README.md"
@@ -46,7 +46,10 @@ dependencies = [
   "composio-core==0.7.0",
   "networkx>=3.4.2",
   "matplotlib>=3.10.0",
-  "boto3>=1.37.1"
+  "boto3>=1.37.1",
+  "scikit-learn>=1.6.1",
+  "numpy>=1.26.4",
+  "pandas>=2.2.3",
 ]
 classifiers = [
   "Programming Language :: Python",
@@ -89,11 +92,7 @@ torch = [
     "torch>=2.6.0",
     "torchvision>=0.21.0",
 ]
-evals = [
-    "scikit-learn>=1.6.1",
-    "numpy>=1.26.4",
-    "pandas>=2.2.3",
-]
+
 
 [tool.uv]
 dev-dependencies = [

{versionhq-1.2.3.4 → versionhq-1.2.3.6}/requirements.txt

@@ -22,3 +22,6 @@ composio-core==0.7.0
 networkx>=3.4.2
 matplotlib>=3.10.0
 boto3>=1.37.1
+scikit-learn>=1.6.1
+numpy>=1.26.4
+pandas>=2.2.3

{versionhq-1.2.3.4 → versionhq-1.2.3.6}/src/versionhq/__init__.py

@@ -32,7 +32,7 @@ from versionhq.agent_network.formation import form_agent_network
 from versionhq.task_graph.draft import workflow
 
 
-__version__ = "1.2.3.4"
+__version__ = "1.2.3.6"
 __all__ = [
     "Agent",
 

{versionhq-1.2.3.4 → versionhq-1.2.3.6}/src/versionhq/agent/TEMPLATES/Backstory.py

@@ -1,4 +1,4 @@
 BACKSTORY_FULL="""You are an expert {role} highly skilled in {skills}. You have abilities to query relevant information from the given knowledge sources and use tools such as {tools}. Leveraging these, you will identify competitive solutions to achieve the following goal: {goal}."""
 
 
-BACKSTORY_SHORT="""You are an expert {role} with relevant skillsets and abilities to query relevant information from the given knowledge sources. Leveraging these, you will identify competitive solutions to achieve the following goal: {goal}."""
+BACKSTORY_SHORT="""You are an expert {role} with relevant skills and abilities to query relevant information from the given knowledge sources. Leveraging these, you will identify competitive solutions to achieve the following goal: {goal}."""

{versionhq-1.2.3.4 → versionhq-1.2.3.6}/src/versionhq/agent/model.py

@@ -32,13 +32,15 @@ class Agent(BaseModel):
     _request_within_rpm_limit: Any = PrivateAttr(default=None)
     _times_executed: int = PrivateAttr(default=0)
     _logger_config: Dict[str, Any] = PrivateAttr(default=dict(verbose=True, info_file_save=True))
+
+    api_key: Optional[str] = Field(default=None)
     config: Optional[Dict[str, Any]] = Field(default=None, exclude=True, description="values to add to the Agent class")
 
     id: UUID4 = Field(default_factory=uuid.uuid4, frozen=True)
     role: str = Field(description="required. agent's role")
     goal: Optional[str] = Field(default=None)
     backstory: Optional[str] = Field(default=None, description="developer prompt to the llm")
-    skillsets: Optional[List[str]] = Field(default_factory=list)
+    skills: Optional[List[str]] = Field(default_factory=list, description="list up the agent's tangible skills in natural language")
     tools: Optional[List[Any]] = Field(default_factory=list)
 
     # knowledge
@@ -149,8 +151,12 @@ class Agent(BaseModel):
                                 setattr(tool, k, v)
                         tool_list.append(tool)
 
+                case callable():
+                    tool = Tool(func=item)
+                    tool_list.append(tool)
+
                 case _:
-                    if item.__base__ == BaseTool or item.__base__ == RagTool or item.__base__ == Tool:
+                    if hasattr(item, "__base__") and (item.__base__ == BaseTool or item.__base__ == RagTool or item.__base__ == Tool):
                         tool_list.append(item)
                     else:
                         Logger(**self._logger_config, filename=self.key).log(level="error", message=f"Tool {str(item)} is missing a function.", color="red")
@@ -169,12 +175,12 @@ class Agent(BaseModel):
         if self.backstory is None:
             from versionhq.agent.TEMPLATES.Backstory import BACKSTORY_FULL, BACKSTORY_SHORT
             backstory = ""
-            skills = ", ".join([item for item in self.skillsets]) if self.skillsets else ""
+            skills = ", ".join([item for item in self.skills]) if self.skills else ""
             tools = ", ".join([item.name for item in self.tools if hasattr(item, "name") and item.name is not None]) if self.tools else ""
             role = self.role.lower()
             goal = self.goal.lower() if self.goal else ""
 
-            if self.tools or self.skillsets:
+            if self.tools or self.skills:
                 backstory = BACKSTORY_FULL.format(role=role, goal=goal, skills=skills, tools=tools)
             else:
                 backstory = BACKSTORY_SHORT.format(role=role, goal=goal)
@@ -371,7 +377,7 @@ class Agent(BaseModel):
         messages = []
         messages.append({ "role": "user", "content": prompts })
         if self.use_developer_prompt:
-            messages.append({ "role": "system", "content": self.backstory })
+            messages.append({ "role": "developer", "content": self.backstory })
 
         try:
             if self._rpm_controller and self.max_rpm:

{versionhq-1.2.3.4 → versionhq-1.2.3.6}/src/versionhq/llm/model.py

@@ -8,8 +8,7 @@ from contextlib import contextmanager
 from typing import Any, Dict, List, Optional
 from typing_extensions import Self
 
-import litellm
-from litellm import JSONSchemaValidationError
+from litellm import JSONSchemaValidationError, get_supported_openai_params
 from pydantic import BaseModel, Field, PrivateAttr, model_validator, ConfigDict
 
 from versionhq.llm.llm_vars import LLM_CONTEXT_WINDOW_SIZES, MODELS, PARAMS, PROVIDERS, ENDPOINT_PROVIDERS, ENV_VARS
@@ -48,6 +47,7 @@ class FilteredStream:
 @contextmanager
 def suppress_warnings():
     with warnings.catch_warnings():
+        import litellm
         litellm.set_verbose = False
         warnings.filterwarnings(action="ignore")
         old_stdout = sys.stdout
@@ -94,8 +94,6 @@ class LLM(BaseModel):
 
     model_config = ConfigDict(extra="allow")
 
-    litellm.drop_params = True
-    litellm.set_verbose = True
     os.environ['LITELLM_LOG'] = 'DEBUG'
 
 
@@ -104,6 +102,8 @@ class LLM(BaseModel):
         """
         Validate the given model, provider, interface provider.
         """
+        import litellm
+        litellm.drop_params = True
 
         self._init_model_name = self.model
 
@@ -180,6 +180,9 @@ class LLM(BaseModel):
         """
         Set up valid config params after setting up a valid model, provider, interface provider names.
         """
+        import litellm
+        litellm.drop_params = True
+
         self._tokens = 0
 
         if self.callbacks:
@@ -202,11 +205,10 @@ class LLM(BaseModel):
         """
         Returns valid params incl. model + litellm original params) from the given config dict.
         """
-
         valid_config, valid_keys = dict(), list()
 
         if self.model:
-            valid_keys = litellm.get_supported_openai_params(
+            valid_keys = get_supported_openai_params(
                 model=self.model, custom_llm_provider=self.endpoint_provider, request_type="chat_completion"
             )
 
@@ -257,7 +259,7 @@ class LLM(BaseModel):
     def _supports_function_calling(self) -> bool:
         try:
             if self.model:
-                params = litellm.get_supported_openai_params(model=self.model)
+                params = get_supported_openai_params(model=self.model)
                 return "response_format" in params if params else False
             else:
                 return False
@@ -267,7 +269,8 @@ class LLM(BaseModel):
 
 
     def _supports_stop_words(self) -> bool:
-        supported_params = litellm.get_supported_openai_params(model=self.model, custom_llm_provider=self.endpoint_provider)
+        import litellm
+        supported_params = get_supported_openai_params(model=self.model, custom_llm_provider=self.endpoint_provider)
         return "stop" in supported_params if supported_params else False
 
 
@@ -279,6 +282,7 @@ class LLM(BaseModel):
 
 
     def _set_callbacks(self, callbacks: List[Any]):
+        import litellm
         callback_types = [type(callback) for callback in callbacks]
         for callback in litellm.success_callback[:]:
             if type(callback) in callback_types:
@@ -302,8 +306,9 @@ class LLM(BaseModel):
         """
         Execute LLM based on the agent's params and model params.
         """
-
+        import litellm
         litellm.drop_params = True
+        litellm.set_verbose = True
 
         with suppress_warnings():
             if len(self.callbacks) > 0:

{versionhq-1.2.3.4 → versionhq-1.2.3.6}/src/versionhq.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: versionhq
-Version: 1.2.3.4
+Version: 1.2.3.6
 Summary: An agentic orchestration framework for building agent networks that handle task automation.
 Author-email: Kuriko Iwai <kuriko@versi0n.io>
 License: MIT License
@@ -68,6 +68,9 @@ Requires-Dist: composio-core==0.7.0
 Requires-Dist: networkx>=3.4.2
 Requires-Dist: matplotlib>=3.10.0
 Requires-Dist: boto3>=1.37.1
+Requires-Dist: scikit-learn>=1.6.1
+Requires-Dist: numpy>=1.26.4
+Requires-Dist: pandas>=2.2.3
 Provides-Extra: docling
 Requires-Dist: docling>=2.25.2; extra == "docling"
 Provides-Extra: mem0ai
@@ -82,10 +85,6 @@ Requires-Dist: sec-api>=1.0.28; extra == "tools"
 Provides-Extra: torch
 Requires-Dist: torch>=2.6.0; extra == "torch"
 Requires-Dist: torchvision>=0.21.0; extra == "torch"
-Provides-Extra: evals
-Requires-Dist: scikit-learn>=1.6.1; extra == "evals"
-Requires-Dist: numpy>=1.26.4; extra == "evals"
-Requires-Dist: pandas>=2.2.3; extra == "evals"
 
 # Overview
 
@@ -143,9 +142,9 @@ Agentic orchestration framework for multi-agent networks and task graphs for com
 
 ## Key Features
 
-`versionhq` is a Python framework for agent networks that handle complex task automation without human interaction.
+`versionhq` is a Python framework designed for automating complex, multi-step tasks using autonomous agent networks.
 
-Agents are model-agnostic, and will improve task output, while optimizing token cost and job latency, by sharing their memory, knowledge base, and RAG tools with other agents in the network.
+Users can either configure their agents and network manually or allow the system to automatically manage the process based on provided task goals.
 
 
 ###  Agent Network
@@ -172,7 +171,7 @@ Each node is triggered by specific events and executed by an assigned agent once
 While the network automatically reconfigures itself, you retain the ability to direct the agents using `should_reform` variable.
 
 
-The following code snippet demonstrates the `TaskGraph` and its visualization, saving the diagram to the `uploads` directory.
+The following code snippet explicitly demonstrates the `TaskGraph` and its visualization, saving the diagram to the `uploads` directory.
 
 ```python
 import versionhq as vhq
@@ -214,15 +213,13 @@ A `TaskGraph` represents tasks as `nodes` and their execution dependencies as `e
 
 `Agent Networks` can handle `TaskGraph` objects by optimizing their formations.
 
-The following example demonstrates a simple concept of a `supervising` agent network handling a task graph with three tasks and one critical edge.
-
 <img src="https://res.cloudinary.com/dfeirxlea/image/upload/v1739337639/pj_m_home/zfg4ccw1m1ww1tpnb0pa.png">
 
 <hr />
 
 ### Optimization
 
-Agents are model-agnostic and can handle multiple tasks, leveraging their own and their peers' knowledge sources, memories, and tools.
+Autonomous agents are model-agnostic and can leverage their own and their peers' knowledge sources, memories, and tools.
 
 Agents are optimized during network formation, but customization is possible before or after.
 
@@ -581,7 +578,9 @@ Common issues and solutions:
 
 * Issues related to dependencies: `rm -rf uv.lock`, `uv cache clean`, `uv venv`, and run `uv pip install -r requirements.txt -v`.
 
-* Issues related to agents and other systems: Check `.logs` directory located in the root directory for error messages and stack traces.
+* Issues related to `torch` installation: Add optional dependencies by `uv add versionhq[torch]`.
+
+* Issues related to agents and other systems: Check `.logs` directory located at the root of the project directory for error messages and stack traces.
 
 * Issues related to `Python quit unexpectedly`: Check [this stackoverflow article](https://stackoverflow.com/questions/59888499/macos-catalina-python-quit-unexpectedly-error).
 

{versionhq-1.2.3.4 → versionhq-1.2.3.6}/src/versionhq.egg-info/requires.txt

@@ -22,15 +22,13 @@ composio-core==0.7.0
 networkx>=3.4.2
 matplotlib>=3.10.0
 boto3>=1.37.1
-
-[docling]
-docling>=2.25.2
-
-[evals]
 scikit-learn>=1.6.1
 numpy>=1.26.4
 pandas>=2.2.3
 
+[docling]
+docling>=2.25.2
+
 [mem0ai]
 mem0ai>=0.1.55
 

{versionhq-1.2.3.4 → versionhq-1.2.3.6}/tests/agent/agent_test.py

@@ -56,11 +56,11 @@ def test_build_agent():
     agent = Agent(
         role="analyst",
         goal="analyze the company's website and retrieve the product overview",
-        skillsets=["financial analysis", "product management", ]
+        skills=["financial analysis", "product management", ]
     )
 
     assert agent.role == "analyst"
-    assert agent.backstory == BACKSTORY_FULL.format(role=agent.role.lower(), goal=agent.goal.lower(), skills=", ".join([item for item in agent.skillsets]), tools="")
+    assert agent.backstory == BACKSTORY_FULL.format(role=agent.role.lower(), goal=agent.goal.lower(), skills=", ".join([item for item in agent.skills]), tools="")
     assert isinstance(agent.llm, LLM)
     assert agent.llm.model == DEFAULT_MODEL_NAME
     # assert agent.llm.api_key == LITELLM_API_KEY
@@ -72,14 +72,14 @@ def test_build_agent_with_llm():
     agent = Agent(
         role="analyst",
         goal="analyze the company's website and retrieve the product overview",
-        skillsets=["financial analysis", "product management", ],
+        skills=["financial analysis", "product management", ],
         llm="gpt-4o"
     )
 
     assert agent.role == "analyst"
     assert agent.role in agent.backstory
     assert agent.goal in agent.backstory
-    assert [item in agent.backstory for item in agent.skillsets]
+    assert [item in agent.backstory for item in agent.skills]
     assert isinstance(agent.llm, LLM)
     assert agent.llm.model == "gpt-4o"
     # assert agent.llm.api_key == LITELLM_API_KEY

{versionhq-1.2.3.4 → versionhq-1.2.3.6}/tests/agent/doc_test.py

@@ -62,7 +62,7 @@ def test_docs_core_agent_c1():
         goal="Coping with price competition in saturated markets"
     )
 
-    assert agent.backstory == "You are an expert marketing analyst with relevant skillsets and abilities to query relevant information from the given knowledge sources. Leveraging these, you will identify competitive solutions to achieve the following goal: coping with price competition in saturated markets."
+    assert agent.backstory == "You are an expert marketing analyst with relevant skills and abilities to query relevant information from the given knowledge sources. Leveraging these, you will identify competitive solutions to achieve the following goal: coping with price competition in saturated markets."
 
 
 def test_docs_core_agent_c2():

{versionhq-1.2.3.4 → versionhq-1.2.3.6}/tests/llm/llm_connection_test.py

@@ -37,7 +37,7 @@ def res_field_task():
     return Task(description="return random values strictly following the given response format.", response_fields=demo_response_fields)
 
 
-def test_con_bedrock(simple_task, tool_task, schema_task, res_field_task):
+def _test_con_bedrock(simple_task, tool_task, schema_task, res_field_task):
     llms_to_test = [
         "bedrock/converse/us.meta.llama3-3-70b-instruct-v1:0",
         "bedrock/us.meta.llama3-2-11b-instruct-v1:0",
@@ -67,7 +67,7 @@ def test_con_bedrock(simple_task, tool_task, schema_task, res_field_task):
         assert [v and type(v) == res_field_task.response_fields[i].data_type for i, (k, v) in enumerate(res_4.json_dict.items())]
 
 
-def test_con_gpt(simple_task, tool_task, schema_task, res_field_task):
+def _test_con_gpt(simple_task, tool_task, schema_task, res_field_task):
     llms_to_test = [
         "gpt-4.5-preview-2025-02-27",
     ]

{versionhq-1.2.3.4 → versionhq-1.2.3.6}/tests/llm/llm_test.py

@@ -1,5 +1,3 @@
-import litellm
-
 from versionhq.llm.llm_vars import MODELS, LLM_CONTEXT_WINDOW_SIZES
 from versionhq.llm.model import LLM, DEFAULT_CONTEXT_WINDOW_SIZE
 
@@ -23,6 +21,8 @@ def test_create_llm_from_valid_name():
             assert llm.context_window_size == int(LLM_CONTEXT_WINDOW_SIZES.get(model_name) *0.75) if LLM_CONTEXT_WINDOW_SIZES.get(model_name) is not None else DEFAULT_CONTEXT_WINDOW_SIZE
             assert llm._supports_function_calling() is not None
             assert llm._supports_stop_words() is not None
+
+            import litellm
             assert litellm.callbacks == [dummy_func,]
 
 
@@ -38,6 +38,8 @@ def test_create_llm_from_invalid_name():
     assert llm.context_window_size == int(128000 * 0.75)
     assert llm._supports_function_calling() == True
     assert llm._supports_stop_words() == True
+
+    import litellm
     assert litellm.callbacks == [dummy_func,]
 
 
@@ -49,6 +51,8 @@ def test_create_llm_from_provider():
     assert llm.context_window_size == int(LLM_CONTEXT_WINDOW_SIZES.get(llm.model) *0.75)
     assert llm._supports_function_calling() == True
     assert llm._supports_stop_words() == True
+
+    import litellm
     assert litellm.callbacks == [dummy_func,]