universal-mcp-agents 0.1.23rc9__tar.gz → 0.1.23rc10__tar.gz

{universal_mcp_agents-0.1.23rc9 → universal_mcp_agents-0.1.23rc10}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: universal-mcp-agents
-Version: 0.1.23rc9
+Version: 0.1.23rc10
 Summary: Add your description here
 Project-URL: Homepage, https://github.com/universal-mcp/applications
 Project-URL: Repository, https://github.com/universal-mcp/applications

{universal_mcp_agents-0.1.23rc9 → universal_mcp_agents-0.1.23rc10}/pyproject.toml

@@ -6,7 +6,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "universal-mcp-agents"
-version = "0.1.23-rc9"
+version = "0.1.23-rc10"
 description = "Add your description here"
 readme = "README.md"
 authors = [

{universal_mcp_agents-0.1.23rc9 → universal_mcp_agents-0.1.23rc10}/src/universal_mcp/applications/llm/app.py

@@ -38,7 +38,7 @@ class LlmApp(BaseApplication):
         """Initialize the LLMApp."""
         super().__init__(name="llm")
 
-    async def generate_text(
+    def generate_text(
         self,
         task: str,
         context: str | list[str] | dict[str, str] = "",
@@ -92,10 +92,10 @@ class LlmApp(BaseApplication):
         full_prompt = f"{prompt}\n\nContext:\n{context_str}\n\n"
 
         model = load_chat_model("azure/gpt-5-mini")
-        response = await model.with_retry(stop_after_attempt=MAX_RETRIES).ainvoke(full_prompt)
+        response = model.with_retry(stop_after_attempt=MAX_RETRIES).invoke(full_prompt, stream=False)
         return str(response.content)
 
-    async def classify_data(
+    def classify_data(
         self,
         classification_task_and_requirements: str,
         context: Any | list[Any] | dict[str, Any],
@@ -165,10 +165,10 @@ class LlmApp(BaseApplication):
             response_format=ClassificationResult,  # Auto-selects ProviderStrategy
         )
 
-        result = await agent.ainvoke({"messages": [{"role": "user", "content": prompt}]})
+        result = agent.invoke({"messages": [{"role": "user", "content": prompt}]}, stream=False)
         return result["structured_response"].model_dump()
 
-    async def extract_data(
+    def extract_data(
         self,
         extraction_task: str,
         source: Any | list[Any] | dict[str, Any],
@@ -231,14 +231,14 @@ class LlmApp(BaseApplication):
 
         model = load_chat_model("azure/gpt-5-mini", temperature=0)
 
-        response = await (
+        response = (
             model.with_structured_output(schema=output_schema, method="json_mode")
             .with_retry(stop_after_attempt=MAX_RETRIES)
-            .ainvoke(prompt)
+            .invoke(prompt, stream=False)
         )
         return cast(dict[str, Any], response)
 
-    async def call_llm(
+    def call_llm(
         self,
         task_instructions: str,
         context: Any | list[Any] | dict[str, Any],
@@ -289,7 +289,7 @@ class LlmApp(BaseApplication):
             tools=[],
             response_format=output_schema,
         )
-        result = await agent.ainvoke({"messages": [{"role": "user", "content": prompt}]})
+        result = agent.invoke({"messages": [{"role": "user", "content": prompt}]}, stream=False)
         return result["structured_response"]
 
     def list_tools(self):

{universal_mcp_agents-0.1.23rc9 → universal_mcp_agents-0.1.23rc10}/uv.lock

@@ -4902,7 +4902,7 @@ wheels = [
 
 [[package]]
 name = "universal-mcp-agents"
-version = "0.1.23rc8"
+version = "0.1.23rc9"
 source = { editable = "." }
 dependencies = [
     { name = "cloudpickle" },