versionhq 1.2.1.2__py3-none-any.whl → 1.2.1.3__py3-none-any.whl

versionhq/__init__.py

@@ -30,7 +30,7 @@ from versionhq.memory.model import ShortTermMemory,LongTermMemory, UserMemory, M
 from versionhq.task.formation import form_agent_network
 
 
-__version__ = "1.2.1.2"
+__version__ = "1.2.1.3"
 __all__ = [
     "Agent",
 

versionhq/agent/model.py

@@ -517,12 +517,11 @@ class Agent(BaseModel):
 
 
 
-    def execute_task(self, task, context: Optional[str] = None, task_tools: Optional[List[Tool | ToolSet]] = list()) -> str:
+    def execute_task(self, task, context: Optional[Any] = None, task_tools: Optional[List[Tool | ToolSet]] = list()) -> str:
         """
-        Execute the task and return the response in string.
-        The agent utilizes the tools in task or their own tools if the task.can_use_agent_tools is True.
-        The agent must consider the context to excute the task as well when it is given.
+        Format a task prompt, adding context from knowledge and memory (if given), and invoke LLM.
         """
+
         from versionhq.task.model import Task
         from versionhq.knowledge._utils import extract_knowledge_context
 
@@ -532,9 +531,7 @@ class Agent(BaseModel):
         if self.max_rpm and self._rpm_controller:
             self._rpm_controller._reset_request_count()
 
-        task_prompt = task.prompt(model_provider=self.llm.provider)
-        if context is not task.prompt_context:
-            task_prompt += context
+        task_prompt = task._prompt(model_provider=self.llm.provider, context=context)
 
         if self._knowledge:
             agent_knowledge = self._knowledge.query(query=[task_prompt,], limit=5)
@@ -547,7 +544,9 @@ class Agent(BaseModel):
             contextual_memory = ContextualMemory(
                 memory_config=self.memory_config, stm=self.short_term_memory, ltm=self.long_term_memory, um=self.user_memory
             )
-            memory = contextual_memory.build_context_for_task(task=task, context=context)
+            context_str = task._draft_context_prompt(context=context)
+            query = f"{task.description} {context_str}".strip()
+            memory = contextual_memory.build_context_for_task(query=query)
             if memory.strip() != "":
                 task_prompt += memory.strip()
 

versionhq/memory/contextual_memory.py

@@ -23,14 +23,11 @@ class ContextualMemory:
         self.um = um
 
 
-    def build_context_for_task(self, task, context: List[Any] | str) -> str:
+    def build_context_for_task(self, query: str = None) -> str:
         """
         Automatically builds a minimal, highly relevant set of contextual information for a given task.
         """
-
-        query = f"{task.description} {context}".strip()
-
-        if query == "":
+        if not query:
             return ""
 
         context = []

versionhq/task/formation.py

@@ -82,32 +82,38 @@ def form_agent_network(
         res = vhq_task.execute(agent=vhq_formation_planner, context=context)
         _formation = Formation.SUPERVISING
 
+
         if res.pydantic:
             formation_keys = [k for k, v in Formation._member_map_.items() if k == res.pydantic.formation.upper()]
 
             if formation_keys:
                 _formation = Formation[formation_keys[0]]
 
-            created_agents = [Agent(role=item, goal=item) for item in res.pydantic.agent_roles]
-            created_tasks = [Task(description=item) for item in res.pydantic.task_descriptions]
 
             network_tasks = []
             members = []
             leader = str(res.pydantic.leader_agent)
 
+            created_agents = [Agent(role=item, goal=item) for item in res.pydantic.agent_roles]
+            created_tasks = [Task(description=item) for item in res.pydantic.task_descriptions]
+
+
             for i in range(len(created_agents)):
                 is_manager = bool(created_agents[i].role.lower() == leader.lower())
                 member = Member(agent=created_agents[i], is_manager=is_manager)
 
                 if len(created_tasks) >= i and created_tasks[i]:
                     member.tasks.append(created_tasks[i])
-
                 members.append(member)
 
 
             if len(created_agents) < len(created_tasks):
                 network_tasks.extend(created_tasks[len(created_agents):len(created_tasks)])
 
+            if _formation == Formation.SUPERVISING and not [member for member in members if member.is_manager]:
+                manager = Member(agent=Agent(role=leader, goal=leader), is_manager=True)
+                members.append(manager)
+
             members.sort(key=lambda x: x.is_manager == False)
             network = AgentNetwork(members=members, formation=_formation, network_tasks=network_tasks)
             return network
@@ -138,8 +144,13 @@ def form_agent_network(
             if len(created_agents) < len(created_tasks):
                 network_tasks.extend(created_tasks[len(created_agents):len(created_tasks)])
 
+            if _formation == Formation.SUPERVISING and not [member for member in members if member.is_manager]:
+                member = Member(agent=Agent(role=leader, goal=leader), is_manager=True)
+                members.append(member)
+
             members.sort(key=lambda x: x.is_manager == False)
             network = AgentNetwork(members=members, formation=_formation,  network_tasks=network_tasks)
+
             return network
 
 

versionhq/task/model.py

@@ -196,7 +196,7 @@ class TaskOutput(BaseModel):
         """
         When the task is called as context, return its output in concise string to add it to the prompt
         """
-        return json.dumps(self.json_dict) if self.json_dict else self.raw[0: 127]
+        return json.dumps(self.json_dict) if self.json_dict else self.raw[0: 1024]
 
 
     def evaluate(self, task) -> Evaluation:
@@ -245,7 +245,7 @@ class TaskOutput(BaseModel):
 
 
     @property
-    def json(self) -> Optional[str]:
+    def json_string(self) -> Optional[str]:
         return json.dumps(self.json_dict)
 
 
@@ -259,7 +259,6 @@ class Task(BaseModel):
     """
 
     __hash__ = object.__hash__
-    _logger: Logger = PrivateAttr(default_factory=lambda: Logger(verbose=True))
     _original_description: str = PrivateAttr(default=None)
     _task_output_handler = TaskOutputStorageHandler()
     config: Optional[Dict[str, Any]] = Field(default=None, description="values to set on Task class")
@@ -271,18 +270,17 @@ class Task(BaseModel):
     # output
     pydantic_output: Optional[Type[BaseModel]] = Field(default=None, description="store Pydantic class as structured response format")
     response_fields: Optional[List[ResponseField]] = Field(default_factory=list, description="store list of ResponseField as structured response format")
-    output: Optional[TaskOutput] = Field(default=None, description="store the final task output in TaskOutput class")
 
     # task setup
-    context: Optional[List["Task"]] = Field(default=None, description="other tasks whose outputs should be used as context")
-    prompt_context: Optional[str] = Field(default=None)
+    # context: Optional[List["Task"]] = Field(default=None, description="other tasks whose outputs should be used as context")
+    # prompt_context: Optional[str] = Field(default=None)
 
     # tool usage
     tools: Optional[List[ToolSet | Tool | Any]] = Field(default_factory=list, description="tools that the agent can use aside from their tools")
     can_use_agent_tools: bool = Field(default=False, description="whether the agent can use their own tools when executing the task")
     tool_res_as_final: bool = Field(default=False, description="when set True, tools res will be stored in the `TaskOutput`")
 
-    # execution rules
+    # executing
     execution_type: TaskExecutionType = Field(default=TaskExecutionType.SYNC)
     allow_delegation: bool = Field(default=False, description="ask other agents for help and run the task instead")
     callback: Optional[Callable] = Field(default=None, description="callback to be executed after the task is completed.")
@@ -294,10 +292,11 @@ class Task(BaseModel):
 
     # recording
     processed_agents: Set[str] = Field(default_factory=set, description="store roles of the agents that executed the task")
-    tools_errors: int = 0
+    tool_errors: int = 0
     delegations: int = 0
     latency: int | float = 0 # job latency in sec
     tokens: int = 0 # tokens consumed
+    output: Optional[TaskOutput] = Field(default=None, description="store the final task output in TaskOutput class")
 
 
     @model_validator(mode="before")
@@ -339,13 +338,6 @@ class Task(BaseModel):
         return self
 
 
-    @model_validator(mode="after")
-    def backup_description(self):
-        if self._original_description == None:
-            self._original_description = self.description
-        return self
-
-
     def _draft_output_prompt(self, model_provider: str) -> str:
         """
         Draft prompts on the output format by converting `
@@ -371,35 +363,60 @@ Your response MUST be a valid JSON string that strictly follows the response for
 Response format: {response_format}
 Ref. Output image: {output_formats_to_follow}
 """
-
         else:
             output_prompt = "Return your response as a valid JSON serializable string, enclosed in double quotes. Do not use single quotes, trailing commas, or other non-standard JSON syntax."
 
         return output_prompt
 
 
-    def prompt(self, model_provider: str = None, customer: str = None, product_overview: str = None) -> str:
+    def _draft_context_prompt(self, context: Any) -> str:
         """
-        Format the task prompt and cascade it to the agent.
-        When the task has context, add context prompting of all the tasks in the context.
-        When we have cusotmer/product info, add them to the prompt.
+        Create a context prompt from the given context in any format: a task object, task output object, list, dict.
         """
 
-        output_prompt = self._draft_output_prompt(model_provider=model_provider)
-        task_slices = [self.description, output_prompt,]
+        context_to_add = None
+        if not context:
+            Logger().log(level="error", color="red", message="Missing a context to add to the prompt. We'll return ''.")
+            return context_to_add
+
+        match context:
+            case str():
+                context_to_add = context
+
+            case Task():
+                if not context.output:
+                    res = context.execute()
+                    context_to_add = res.raw
+
+                else:
+                    context_to_add = context.output.raw
+
+            case TaskOutput():
+                context_to_add = context.raw
+
+            case dict():
+                context_to_add = str(context)
+
+            case list():
+                res = ", ".join([self._draft_context_prompt(context=item) for item in context])
+                context_to_add = res
+
+            case _:
+                pass
 
-        if self.context:
-            context_outputs = "\n".join([task.output.context_prompting() if hasattr(task, "output") else "" for task in self.context])
-            task_slices.insert(len(task_slices), f"Consider the following context when responding: {context_outputs}")
+        return context_to_add
 
-        if self.prompt_context:
-            task_slices.insert(len(task_slices), f"Consider the following context when responding: {self.prompt_context}")
 
-        if customer:
-            task_slices.insert(len(task_slices), f"Customer to address: {customer}")
+    def _prompt(self, model_provider: str = None, context: Optional[Any] = None) -> str:
+        """
+        Format the task prompt and cascade it to the agent.
+        """
+        output_prompt = self._draft_output_prompt(model_provider=model_provider)
+        context_prompt = self._draft_context_prompt(context=context) if context else None
+        task_slices = [self.description, output_prompt, ]
 
-        if product_overview:
-            task_slices.insert(len(task_slices), f"Product to promote: {product_overview}")
+        if context_prompt:
+            task_slices.insert(len(task_slices), f"Consider the following context when responding: {context_prompt}")
 
         return "\n".join(task_slices)
 
@@ -449,8 +466,8 @@ Ref. Output image: {output_formats_to_follow}
         """
 
         if raw is None or raw == "":
-            self._logger.log(level="warning", message="The model returned an empty response. Returning an empty dict.", color="yellow")
-            output = { "output": "n.a." }
+            Logger().log(level="warning", message="The model returned an empty response. Returning an empty dict.", color="yellow")
+            output = { "output": "" }
             return output
 
         try:
@@ -498,8 +515,10 @@ Ref. Output image: {output_formats_to_follow}
 
     def interpolate_inputs(self, inputs: Dict[str, Any]) -> None:
         """
-        Interpolate inputs into the task description and expected output.
+        Interpolate inputs into the task description.
         """
+        self._original_description = self.description
+
         if inputs:
             self.description = self._original_description.format(**inputs)
 
@@ -537,17 +556,18 @@ Ref. Output image: {output_formats_to_follow}
                 )
 
         except AttributeError as e:
-            self._logger.log(level="error", message=f"Missing attributes for long term memory: {str(e)}", color="red")
+            Logger().log(level="error", message=f"Missing attributes for long term memory: {str(e)}", color="red")
             pass
 
         except Exception as e:
-            self._logger.log(level="error", message=f"Failed to add to the memory: {str(e)}", color="red")
+            Logger().log(level="error", message=f"Failed to add to the memory: {str(e)}", color="red")
             pass
 
+
     def _build_agent_from_task(self, task_description: str = None) -> InstanceOf["vhq.Agent"]:
         task_description = task_description if task_description else self.description
         if not task_description:
-            self._logger.log(level="error", message="Task is missing the description.", color="red")
+            Logger().log(level="error", message="Task is missing the description.", color="red")
             pass
 
         agent = vhq.Agent(goal=task_description, role=task_description, maxit=1) #! REFINEME
@@ -555,7 +575,9 @@ Ref. Output image: {output_formats_to_follow}
 
 
     # task execution
-    def execute(self, type: TaskExecutionType = None, agent: Optional[Any] = None, context: Optional[str] = None) -> TaskOutput | Future[TaskOutput]:
+    def execute(
+            self, type: TaskExecutionType = None, agent: Optional["vhq.Agent"] = None, context: Optional[Any] = None
+            ) -> TaskOutput | Future[TaskOutput]:
         """
         A main method to handle task execution. Build an agent when the agent is not given.
         """
@@ -572,26 +594,14 @@ Ref. Output image: {output_formats_to_follow}
                 return self._execute_async(agent=agent, context=context)
 
 
-    def _execute_sync(self, agent, context: Optional[str | List[Any]] = None) -> TaskOutput:
-        """
-        Execute the task synchronously.
-        When the task has context, make sure we have executed all the tasks in the context first.
-        """
-
-        if self.context:
-            if isinstance(self.context, list):
-                for task in self.context:
-                    if isinstance(task, Task) and task.output is None:
-                        task._execute_core(agent, context)
 
+    def _execute_sync(self, agent, context: Optional[Any] = None) -> TaskOutput:
+        """Executes the task synchronously."""
         return self._execute_core(agent, context)
 
 
-    def _execute_async(self, agent, context: Optional[str] = None) -> Future[TaskOutput]:
-        """
-        Execute the task asynchronously.
-        """
-
+    def _execute_async(self, agent, context: Optional[Any] = None) -> Future[TaskOutput]:
+        """Executes the task asynchronously."""
         future: Future[TaskOutput] = Future()
         threading.Thread(daemon=True, target=self._execute_task_async, args=(agent, context, future)).start()
         return future
@@ -601,21 +611,19 @@ Ref. Output image: {output_formats_to_follow}
         """
         Executes the task asynchronously with context handling.
         """
-
         result = self._execute_core(agent, context)
         future.set_result(result)
 
 
-    def _execute_core(self, agent, context: Optional[str]) -> TaskOutput:
+    def _execute_core(self, agent, context: Optional[Any]) -> TaskOutput:
         """
         A core method for task execution.
-        Handles 1. agent delegation, 2. tools, 3. context to add to the prompt, and 4. callbacks
+        Handles 1. agent delegation, 2. tools, 3. context to add to the prompt, and 4. callbacks.
         """
 
         from versionhq.agent.model import Agent
         from versionhq.agent_network.model import AgentNetwork
 
-        self.prompt_context = context
         task_output: InstanceOf[TaskOutput] = None
         raw_output: str = None
         tool_output: str | list = None
@@ -719,52 +727,3 @@ Task ID: {str(self.id)}
 "Description": {self.description}
 "Tools": {", ".join([tool.name for tool in self.tools])}
         """
-
-
-
-# class ConditionalTask(Task):
-#     """
-#     A task that can be conditionally executed based on the output of another task.
-#     When the `condition` return True, execute the task, else skipped with `skipped task output`.
-#     """
-
-#     condition: Callable[[TaskOutput], bool] = Field(
-#         default=None,
-#         description="max. number of retries for an agent to execute a task when an error occurs",
-#     )
-
-
-#     def __init__(self, condition: Callable[[Any], bool], **kwargs):
-#         super().__init__(**kwargs)
-#         self.condition = condition
-
-
-#     def should_execute(self, context: TaskOutput) -> bool:
-#         """
-#         Decide whether the conditional task should be executed based on the provided context.
-#         Return `True` if it should be executed.
-#         """
-#         return self.condition(context)
-
-
-#     def get_skipped_task_output(self):
-#         return TaskOutput(task_id=self.id, raw="", pydantic=None, json_dict={})
-
-
-#     def _handle_conditional_task(self, task_outputs: List[TaskOutput], task_index: int, was_replayed: bool) -> Optional[TaskOutput]:
-#         """
-#         When the conditional task should be skipped, return `skipped_task_output` as task_output else return None
-#         """
-
-#         previous_output = task_outputs[task_index - 1] if task_outputs and len(task_outputs) > 1  else None
-
-#         if previous_output and not self.should_execute(previous_output):
-#             self._logger.log(level="warning", message=f"Skipping conditional task: {self.description}", color="yellow")
-#             skipped_task_output = self.get_skipped_task_output()
-#             self.output = skipped_task_output
-
-#             if not was_replayed:
-#                 self._store_execution_log(self, task_index=task_index, was_replayed=was_replayed, inputs={})
-#             return skipped_task_output
-
-#         return None

{versionhq-1.2.1.2.dist-info → versionhq-1.2.1.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: versionhq
-Version: 1.2.1.2
+Version: 1.2.1.3
 Summary: An agentic orchestration framework for building agent networks that handle task automation.
 Author-email: Kuriko Iwai <kuriko@versi0n.io>
 License: MIT License
@@ -89,7 +89,7 @@ Requires-Dist: pygraphviz>=1.14; extra == "pygraphviz"
 ![pyenv ver](https://img.shields.io/badge/pyenv-2.5.0-orange)
 
 
-A Python framework for agentic orchestration that handles complex task automation without human interaction.
+Agentic orchestration framework for multi-agent networks and task graph for complex task automation.
 
 **Visit:**
 
@@ -188,7 +188,14 @@ task_graph.add_dependency(
    type=vhq.DependencyType.FINISH_TO_FINISH, lag=1, required=False, weight=3
 )
 
+# To visualize the graph:
 task_graph.visualize()
+
+# To start executing nodes:
+latest_output, outputs = task_graph.activate()
+
+assert isinstance(last_task_output, vhq.TaskOutput)
+assert [k in task_graph.nodes.keys() and v and isinstance(v, vhq.TaskOutput) for k, v in outputs.items()]
 ```
 
 <hr />
@@ -346,34 +353,34 @@ Tasks can be delegated to a manager, peers within the agent network, or a comple
 
 ## Technologies Used
 
-**Task Graph**
-
-* [NetworkX](https://networkx.org/documentation/stable/reference/introduction.html): A Python package to analyze, create, and manipulate complex graph networks.
-* [Matplotlib](https://matplotlib.org/stable/index.html): Visualization library
-* [Graphviz](https://graphviz.org/about/): Graph visualization software
-
-
 **Schema, Data Validation**
 
 * [Pydantic](https://docs.pydantic.dev/latest/): Data validation and serialization library for Python.
 * [Upstage](https://console.upstage.ai/docs/getting-started/overview): Document processer for ML tasks. (Use `Document Parser API` to extract data from documents)
 * [Docling](https://ds4sd.github.io/docling/): Document parsing
 
+**Workflow, Task Graph**
 
-**Storage**
-
-* [mem0ai](https://docs.mem0.ai/quickstart#install-package): Agents' memory storage and management.
-* [Chroma DB](https://docs.trychroma.com/): Vector database for storing and querying usage data.
-* [SQLite](https://www.sqlite.org/docs.html): C-language library to implements a small SQL database engine.
+* [NetworkX](https://networkx.org/documentation/stable/reference/introduction.html): A Python package to analyze, create, and manipulate complex graph networks.
+* [Matplotlib](https://matplotlib.org/stable/index.html): For graph visualization.
+* [Graphviz](https://graphviz.org/about/): For graph visualization.
 
-**LLM Integration**
+**LLM Curation**
 
-* [LiteLLM](https://docs.litellm.ai/docs/providers): Integration to diverse LLMs
+* [LiteLLM](https://docs.litellm.ai/docs/providers): LLM orchestration platform
 
 **Tools**
 
 * [Composio](https://composio.dev/): Conect RAG agents with external tools, Apps, and APIs to perform actions and receive triggers. We use [tools](https://composio.dev/tools) and [RAG tools](https://app.composio.dev/app/ragtool) from Composio toolset.
 
+
+**Storage**
+
+* [mem0ai](https://docs.mem0.ai/quickstart#install-package): Agents' memory storage and management.
+* [Chroma DB](https://docs.trychroma.com/): Vector database for storing and querying usage data.
+* [SQLite](https://www.sqlite.org/docs.html): C-language library to implements a small SQL database engine.
+
+
 **Deployment**
 
 * **Python**: Primary programming language. v3.12.x is recommended

{versionhq-1.2.1.2.dist-info → versionhq-1.2.1.3.dist-info}/RECORD

@@ -1,4 +1,4 @@
-versionhq/__init__.py,sha256=nv3StURywg_U4o7_lKbqPuby1xjSphOkY9B6Ra5ORHA,2817
+versionhq/__init__.py,sha256=WBEWODzBMhY4LEtXSvUkIixwfk0-xvGLJSID6O_xVZY,2817
 versionhq/_utils/__init__.py,sha256=dzoZr4cBlh-2QZuPzTdehPUCe9lP1dmRtauD7qTjUaA,158
 versionhq/_utils/i18n.py,sha256=TwA_PnYfDLA6VqlUDPuybdV9lgi3Frh_ASsb_X8jJo8,1483
 versionhq/_utils/logger.py,sha256=IxSlr2Vi7AXaxj5Fuy8LRzEovaIFVwcbWTgJnASsHN8,3155
@@ -7,7 +7,7 @@ versionhq/_utils/usage_metrics.py,sha256=NXF18dn5NNvGK7EsQ4AAghpR8ppYOjMx6ABenLL
 versionhq/_utils/vars.py,sha256=bZ5Dx_bFKlt3hi4-NNGXqdk7B23If_WaTIju2fiTyPQ,57
 versionhq/agent/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 versionhq/agent/inhouse_agents.py,sha256=snDtgDmvZB2bZKH_RTcz5uFOMl3MTjLJwTQBebFt8hk,2532
-versionhq/agent/model.py,sha256=HiEQulVjXx6fWGaP-DBWmiZv5uwXFPd4S8jyBU3ydm0,25442
+versionhq/agent/model.py,sha256=L21COegL50x1sl-1gZqYnICx8Gm276J0GnQx6IthXf8,25343
 versionhq/agent/parser.py,sha256=riG0dkdQCxH7uJ0AbdVdg7WvL0BXhUgJht0VtQvxJBc,4082
 versionhq/agent/rpm_controller.py,sha256=grezIxyBci_lDlwAlgWFRyR5KOocXeOhYkgN02dNFNE,2360
 versionhq/agent/TEMPLATES/Backstory.py,sha256=IAhGnnt6VUMe3wO6IzeyZPDNu7XE7Uiu3VEXUreOcKs,532
@@ -33,7 +33,7 @@ versionhq/llm/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 versionhq/llm/llm_vars.py,sha256=wjQK20cKvph6Vq1v71o4d16zBGcHlwq0bzOT_zWno7w,7041
 versionhq/llm/model.py,sha256=wlzDUMEyIOm808d1vzqu9gmbB4ch-s_EUvwFR60gR80,17177
 versionhq/memory/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-versionhq/memory/contextual_memory.py,sha256=tCsOOAUnfrOL7YiakqGoi3uShzzS870TmGnlGd3z_A4,3556
+versionhq/memory/contextual_memory.py,sha256=WeDujcEp4oud30OusXSPPNrMEQP-vGrt1mcfYopQruU,3483
 versionhq/memory/model.py,sha256=MPO8dDP5eAuk9td6bMOq5j2huLzCADFJDrTujHhcWQY,8150
 versionhq/storage/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 versionhq/storage/base.py,sha256=p-Jas0fXQan_qotnRD6seQxrT2lj-uw9-SmHQhdppcs,355
@@ -44,10 +44,10 @@ versionhq/storage/task_output_storage.py,sha256=E1t_Fkt78dPYIOl3MP7LfQ8oGtjlzxBu
 versionhq/storage/utils.py,sha256=ByYXPoEIGJYLUqz-DWjbCAnneNrH1otiYbp12SCILpM,747
 versionhq/task/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 versionhq/task/evaluate.py,sha256=WdUgjbZL62XrxyWe5MTz29scfzwmuAHGxJ7GvAB8Fmk,3954
-versionhq/task/formation.py,sha256=gBrFmMHvoyqdK9wEa9aBkZImEQV3GMDDX6-RDpw0pEo,6421
+versionhq/task/formation.py,sha256=qkAJ1ToeOHQFR3hqC8nJyU6msftPgm8yEPTrFfz77OA,6906
 versionhq/task/formatter.py,sha256=N8Kmk9vtrMtBdgJ8J7RmlKNMdZWSmV8O1bDexmCWgU0,643
 versionhq/task/log_handler.py,sha256=LT7YnO7gcPR9IZS7eRvMjnHh8crMBFtqduxd8dxIbkk,1680
-versionhq/task/model.py,sha256=H7ZYZpp_1mrw_MlKjLr6gMJztqyhPkdeoolIzZSF68Y,30756
+versionhq/task/model.py,sha256=i-3ABpuE2a5Gp6-jalU0BOWrbolP90EnACDgdIcNAvQ,28722
 versionhq/task/structured_response.py,sha256=4q-hQPu7oMMHHXEzh9YW4SJ7N5eCZ7OfZ65juyl_jCI,5000
 versionhq/task/TEMPLATES/Description.py,sha256=V-4kh8xpQTKOcDMi2xnuP-fcNk6kuoz1_5tYBlDLQWQ,420
 versionhq/task_graph/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -59,8 +59,8 @@ versionhq/tool/composio_tool_vars.py,sha256=FvBuEXsOQUYnN7RTFxT20kAkiEYkxWKkiVtg
 versionhq/tool/decorator.py,sha256=C4ZM7Xi2gwtEMaSeRo-geo_g_MAkY77WkSLkAuY0AyI,1205
 versionhq/tool/model.py,sha256=PO4zNWBZcJhYVur381YL1dy6zqurio2jWjtbxOxZMGI,12194
 versionhq/tool/tool_handler.py,sha256=2m41K8qo5bGCCbwMFferEjT-XZ-mE9F0mDUOBkgivOI,1416
-versionhq-1.2.1.2.dist-info/LICENSE,sha256=cRoGGdM73IiDs6nDWKqPlgSv7aR4n-qBXYnJlCMHCeE,1082
-versionhq-1.2.1.2.dist-info/METADATA,sha256=SjmFRj8FztRldIYEtsObQlY5XTk6mWL-zmr6sCaSWAY,22116
-versionhq-1.2.1.2.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
-versionhq-1.2.1.2.dist-info/top_level.txt,sha256=DClQwxDWqIUGeRJkA8vBlgeNsYZs4_nJWMonzFt5Wj0,10
-versionhq-1.2.1.2.dist-info/RECORD,,
+versionhq-1.2.1.3.dist-info/LICENSE,sha256=cRoGGdM73IiDs6nDWKqPlgSv7aR4n-qBXYnJlCMHCeE,1082
+versionhq-1.2.1.3.dist-info/METADATA,sha256=_Ybf1tHBU0lp8W77c7aswPZvr5DCPgx13vqoLmftGBc,22374
+versionhq-1.2.1.3.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
+versionhq-1.2.1.3.dist-info/top_level.txt,sha256=DClQwxDWqIUGeRJkA8vBlgeNsYZs4_nJWMonzFt5Wj0,10
+versionhq-1.2.1.3.dist-info/RECORD,,