semantio 0.0.2__py3-none-any.whl → 0.0.3__py3-none-any.whl

semantio/agent.py

@@ -21,24 +21,24 @@ import os
 logging.basicConfig(level=logging.INFO)
 logger = logging.getLogger(__name__)
 
-class Assistant(BaseModel):
+class Agent(BaseModel):
     # -*- Agent settings
-    name: Optional[str] = Field(None, description="Name of the assistant.")
-    description: Optional[str] = Field(None, description="Description of the assistant's role.")
-    instructions: Optional[List[str]] = Field(None, description="List of instructions for the assistant.")
+    name: Optional[str] = Field(None, description="Name of the agent.")
+    description: Optional[str] = Field(None, description="Description of the agent's role.")
+    instructions: Optional[List[str]] = Field(None, description="List of instructions for the agent.")
     model: Optional[str] = Field(None, description="This one is not in the use.")
     show_tool_calls: bool = Field(False, description="Whether to show tool calls in the response.")
     markdown: bool = Field(False, description="Whether to format the response in markdown.")
-    tools: Optional[List[BaseTool]] = Field(None, description="List of tools available to the assistant.")
-    user_name: Optional[str] = Field("User", description="Name of the user interacting with the assistant.")
-    emoji: Optional[str] = Field(":robot:", description="Emoji to represent the assistant in the CLI.")
+    tools: Optional[List[BaseTool]] = Field(None, description="List of tools available to the agent.")
+    user_name: Optional[str] = Field("User", description="Name of the user interacting with the agent.")
+    emoji: Optional[str] = Field(":robot:", description="Emoji to represent the agent in the CLI.")
     rag: Optional[RAG] = Field(None, description="RAG instance for context retrieval.")
     knowledge_base: Optional[Any] = Field(None, description="Knowledge base for domain-specific information.")
     llm: Optional[str] = Field(None, description="The LLM provider to use (e.g., 'groq', 'openai', 'anthropic').")
     llm_model: Optional[str] = Field(None, description="The specific model to use for the LLM provider.")
     llm_instance: Optional[BaseLLM] = Field(None, description="The LLM instance to use.")
     json_output: bool = Field(False, description="Whether to format the response as JSON.")
-    api: bool = Field(False, description="Whether to generate an API for the assistant.")
+    api: bool = Field(False, description="Whether to generate an API for the agent.")
     api_config: Optional[Dict] = Field(
         None,
         description="Configuration for the API (e.g., host, port, authentication).",
@@ -46,7 +46,7 @@ class Assistant(BaseModel):
     api_generator: Optional[Any] = Field(None, description="The API generator instance.")
     expected_output: Optional[Union[str, Dict]] = Field(None, description="The expected format or structure of the output.")
     semantic_model: Optional[Any] = Field(None, description="SentenceTransformer model for semantic matching.")
-    team: Optional[List['Assistant']] = Field(None, description="List of assistants in the team.")
+    team: Optional[List['Agent']] = Field(None, description="List of assistants in the team.")
     auto_tool: bool = Field(False, description="Whether to automatically detect and call tools.")
     
     # Allow arbitrary types
@@ -125,7 +125,7 @@ class Assistant(BaseModel):
             try:
                 # Import the module
                 module_name = file.stem
-                module = importlib.import_module(f"hashai.tools.{module_name}")
+                module = importlib.import_module(f"semantio.tools.{module_name}")
 
                 # Find all classes that inherit from BaseTool
                 for name, obj in module.__dict__.items():
@@ -197,7 +197,7 @@ class Assistant(BaseModel):
         model_to_use = self.llm_model or default_model
 
         # Dynamically import and initialize the LLM class
-        module_name = f"hashai.llm.{llm_provider}"
+        module_name = f"semantio.llm.{llm_provider}"
         llm_module = importlib.import_module(module_name)
         llm_class = getattr(llm_module, llm_class_name)
         self.llm_instance = llm_class(model=model_to_use, api_key=api_key)
@@ -214,10 +214,10 @@ class Assistant(BaseModel):
         stream: bool = False,
         markdown: bool = False,
         tools: Optional[List[BaseTool]] = None,
-        team: Optional[List['Assistant']] = None,
+        team: Optional[List['Agent']] = None,
         **kwargs,
     ) -> Union[str, Dict]:  # Add return type hint
-        """Print the assistant's response to the console and return it."""
+        """Print the agent's response to the console and return it."""
 
         if stream:
             # Handle streaming response
@@ -234,14 +234,14 @@ class Assistant(BaseModel):
 
 
     def _stream_response(self, message: str, markdown: bool = False, **kwargs) -> Iterator[str]:
-        """Stream the assistant's response."""
+        """Stream the agent's response."""
         # Simulate streaming by yielding chunks of the response
         response = self._generate_response(message, markdown=markdown, **kwargs)
         for chunk in response.split():
             yield chunk + " "
 
     def register_tool(self, tool: BaseTool):
-        """Register a tool for the assistant."""
+        """Register a tool for the agent."""
         if self.tools is None:
             self.tools = []
         self.tools.append(tool)
@@ -256,7 +256,7 @@ class Assistant(BaseModel):
 
         # Create a prompt for the LLM
         prompt = f"""
-        You are an AI assistant that helps users by selecting the most appropriate tool to answer their query. Below is a list of available tools and their functionalities:
+        You are an AI agent that helps users by selecting the most appropriate tool to answer their query. Below is a list of available tools and their functionalities:
 
         {self._get_tool_descriptions()}
 
@@ -290,7 +290,7 @@ class Assistant(BaseModel):
         """
         # Create a prompt for the LLM to analyze the query and select tools
         prompt = f"""
-        You are an AI assistant that helps analyze user queries and select the most appropriate tools.
+        You are an AI agent that helps analyze user queries and select the most appropriate tools.
         Below is a list of available tools and their functionalities:
 
         {self._get_tool_descriptions()}
@@ -324,8 +324,8 @@ class Assistant(BaseModel):
             return []
     
 
-    def _generate_response(self, message: str, markdown: bool = False, tools: Optional[List[BaseTool]] = None, team: Optional[List['Assistant']] = None, **kwargs) -> str:
-        """Generate the assistant's response, including tool execution and context retrieval."""
+    def _generate_response(self, message: str, markdown: bool = False, tools: Optional[List[BaseTool]] = None, team: Optional[List['Agent']] = None, **kwargs) -> str:
+        """Generate the agent's response, including tool execution and context retrieval."""
         # Use the specified tools or team if provided
         if tools is not None:
             self.tools = tools
@@ -435,12 +435,12 @@ class Assistant(BaseModel):
         # Combine all responses into a single string
         return "\n\n".join(responses)
     
-    def _generate_team_response(self, message: str, team: List['Assistant'], markdown: bool = False, **kwargs) -> str:
+    def _generate_team_response(self, message: str, team: List['Agent'], markdown: bool = False, **kwargs) -> str:
         """Generate a response using a team of assistants."""
         responses = []
-        for assistant in team:
-            response = assistant.print_response(message, markdown=markdown, **kwargs)
-            responses.append(f"**{assistant.name}:**\n\n{response}")
+        for agent in team:
+            response = agent.print_response(message, markdown=markdown, **kwargs)
+            responses.append(f"**{agent.name}:**\n\n{response}")
         return "\n\n".join(responses)
 
     def _build_prompt(self, message: str, context: Optional[List[Dict]]) -> str:
@@ -578,7 +578,7 @@ class Assistant(BaseModel):
         exit_on: Optional[List[str]] = None,
         **kwargs,
     ):
-        """Run the assistant in a CLI app."""
+        """Run the agent in a CLI app."""
         from rich.prompt import Prompt
 
         if message:
@@ -593,15 +593,15 @@ class Assistant(BaseModel):
             self.print_response(message=message, **kwargs)
 
     def _generate_api(self):
-        """Generate an API for the assistant if api=True."""
+        """Generate an API for the agent if api=True."""
         from .api.api_generator import APIGenerator
         self.api_generator = APIGenerator(self)
-        print(f"API generated for assistant '{self.name}'. Use `.run_api()` to start the API server.")
+        print(f"API generated for agent '{self.name}'. Use `.run_api()` to start the API server.")
 
     def run_api(self):
-        """Run the API server for the assistant."""
+        """Run the API server for the agent."""
         if not hasattr(self, 'api_generator'):
-            raise ValueError("API is not enabled for this assistant. Set `api=True` when initializing the assistant.")
+            raise ValueError("API is not enabled for this agent. Set `api=True` when initializing the agent.")
     
         # Get API configuration
         host = self.api_config.get("host", "0.0.0.0") if self.api_config else "0.0.0.0"

semantio/cli/main.py

@@ -1,7 +1,7 @@
 import argparse
 import warnings
-from hashai.assistant import Assistant
-from hashai.llm import get_llm
+from semantio.agent import Agent
+from semantio.llm import get_llm
 from urllib3.exceptions import NotOpenSSLWarning
 
 # Suppress the NotOpenSSLWarning
@@ -9,7 +9,7 @@ warnings.filterwarnings("ignore", category=NotOpenSSLWarning)
 
 def main():
     parser = argparse.ArgumentParser(description="opAi CLI")
-    parser.add_argument("--message", type=str, required=True, help="Message to send to the assistant")
+    parser.add_argument("--message", type=str, required=True, help="Message to send to the agent")
     parser.add_argument("--provider", type=str, required=True, help="LLM provider (e.g., groq, openai)")
     parser.add_argument("--api-key", type=str, required=True, help="API key for the LLM provider")
     parser.add_argument("--model", type=str, default=None, help="Model name (e.g., mixtral-8x7b-32768)")
@@ -22,9 +22,9 @@ def main():
 
     llm = get_llm(provider=args.provider, **llm_config)
 
-    # Create an assistant
-    assistant = Assistant(model=args.provider, llm=llm)
-    assistant.print_response(args.message)
+    # Create an agent
+    agent = Agent(model=args.provider, llm=llm)
+    agent.print_response(args.message)
     
 
 if __name__ == "__main__":

{semantio-0.0.2.dist-info → semantio-0.0.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: semantio
-Version: 0.0.2
+Version: 0.0.3
 Summary: A powerful SDK for building AI agents
 Home-page: https://github.com/Syenah/semantio
 Author: Rakesh

{semantio-0.0.2.dist-info → semantio-0.0.3.dist-info}/RECORD

@@ -1,12 +1,12 @@
 semantio/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-semantio/agent.py,sha256=mwcKaHy-HY26hxgCRZS5LZw1semlj59qF4uYcLpyqBc,30247
+semantio/agent.py,sha256=plQ4D76cnJ1FaGlEuKDeA53aW_hMDvt5sbmUuTHqvFQ,30143
 semantio/memory.py,sha256=eNAwyAokppHzMcIyFgOw2hT2wnLQBd9GL4T5eallNV4,281
 semantio/rag.py,sha256=ROy3Pa1NURcDs6qQZ8IMoa5Xlzt6I-msEq0C1p8UgB0,472
 semantio/api/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 semantio/api/api_generator.py,sha256=Q-USITEpluRESEaQuOmF7m1vhLKYU9P8eGlQppKT9J4,829
 semantio/api/fastapi_app.py,sha256=DyTgKJKikMe2G6wWmyzo1rBLXQFi8UWWUMY3UGH4f24,2128
 semantio/cli/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-semantio/cli/main.py,sha256=BvY2B2xUSV1Q-4K6eH5P1tJB2Y0aL_rm-1ZW-D0F3eU,1115
+semantio/cli/main.py,sha256=jUvSfehbHWALwracEgBopMIVMraSV9QmDUFfgGcxnP0,1091
 semantio/knowledge_base/__init__.py,sha256=mvp0GFiGSjcxlkaDulAwKOCL9s6gsKTqhPKXF9N3n1g,172
 semantio/knowledge_base/document_loader.py,sha256=nix0yZJ-JJoDbhLkpg5bKDMvNrwykmknI7MRIn0N81k,1910
 semantio/knowledge_base/retriever.py,sha256=XpdzKS1UCncJImVMtG67VXMC7lp2eRzKnShjvktsFMM,1271
@@ -17,7 +17,6 @@ semantio/llm/base_llm.py,sha256=VFl_2S4kqYDuCTWIfWMbKU5aNbVqOCG33E4APOSHF90,668
 semantio/llm/deepseek.py,sha256=oxX-Uw0_lY2sstYs5KGBGFB_hAZUbZomPADdib1mY2M,1100
 semantio/llm/gemini.py,sha256=er3zv1jOvWQBGbPuv4fS4pR_c_abHyhroe-rkXupOO4,1959
 semantio/llm/groq.py,sha256=1AH30paKzDIQjBjWPQPN44QwFHsIOVwI-a587-cDIVc,4285
-semantio/llm/llama.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 semantio/llm/mistral.py,sha256=NpvaB1cE6-jMEBdT0mTf6Ca4Qq2LS8QivDKI6AgdRjE,1061
 semantio/llm/openai.py,sha256=I3ab-d_zFxm-TDhYk6t1PzDtElPJEEQ2eSiARBNIGi4,5174
 semantio/storage/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -28,16 +27,15 @@ semantio/tools/base_tool.py,sha256=xBNSa_8a8WmA4BGRLG2dE7wj9GnBcZo7-P2SyD86GvY,5
 semantio/tools/crypto.py,sha256=mut1ztvpPcUUP3b563dh_FmKtP68KmNis3Qm8WENj8w,5559
 semantio/tools/duckduckgo.py,sha256=6mGn0js0cIsVxQlAgB8AYNLP05H8WmJKnSVosiO9iH0,5034
 semantio/tools/stocks.py,sha256=BVuK61O9OmWQjj0YdiCJY6TzpiFJ_An1UJB2RkDfX2k,5393
-semantio/tools/web_browser.py,sha256=LMwPFTHNTtqCp8MEHVlJJUSJa91vM7MZWIL5RDQKF4U,4980
 semantio/utils/__init__.py,sha256=Lx4X4iJpRhZzRmpQb80XXh5Ve8ZMOkadWAxXSmHpO_8,244
 semantio/utils/config.py,sha256=ZTwUTqxjW3-w94zoU7GzivWyJe0JJGvBfuB4RUOuEs8,1198
 semantio/utils/date_utils.py,sha256=x3oqRGv6ee_KCJ0LvCqqZh_FSgS6YGOHBwZQS4TJetY,1471
 semantio/utils/file_utils.py,sha256=b_cMuJINEGk9ikNuNHSn9lsmICWwvtnCDZ03ndH_S2I,1779
 semantio/utils/logger.py,sha256=TmGbP8BRjLMWjXi2GWzZ0RIXt70x9qX3FuIqghCNlwM,510
 semantio/utils/validation_utils.py,sha256=iwoxEb4Q5ILqV6tbesMjPWPCCoL3AmPLejGUy6q8YvQ,1284
-semantio-0.0.2.dist-info/LICENSE,sha256=teQbWD2Zlcl1_Fo29o2tNbs6G26hbCQiUzds5fQGYlY,1063
-semantio-0.0.2.dist-info/METADATA,sha256=LsBDOM1gPw6rbzZ_-TIf4XJxf3DEWZ7s8KDd8_cDMw0,6800
-semantio-0.0.2.dist-info/WHEEL,sha256=ewwEueio1C2XeHTvT17n8dZUJgOvyCWCt0WVNLClP9o,92
-semantio-0.0.2.dist-info/entry_points.txt,sha256=zbPgevSLwcLpdRHqI_atE8EOt8lK2vRF1AoDflDTo18,53
-semantio-0.0.2.dist-info/top_level.txt,sha256=Yte_6mb-bh-I_lQwMjk1GijZkxPoX4Zmp3kBftC1ZlA,9
-semantio-0.0.2.dist-info/RECORD,,
+semantio-0.0.3.dist-info/LICENSE,sha256=teQbWD2Zlcl1_Fo29o2tNbs6G26hbCQiUzds5fQGYlY,1063
+semantio-0.0.3.dist-info/METADATA,sha256=M5Q-waTknpyWrD_HV9G76jMKgPHPrBBwM5Hl8we4ulo,6800
+semantio-0.0.3.dist-info/WHEEL,sha256=ewwEueio1C2XeHTvT17n8dZUJgOvyCWCt0WVNLClP9o,92
+semantio-0.0.3.dist-info/entry_points.txt,sha256=zbPgevSLwcLpdRHqI_atE8EOt8lK2vRF1AoDflDTo18,53
+semantio-0.0.3.dist-info/top_level.txt,sha256=Yte_6mb-bh-I_lQwMjk1GijZkxPoX4Zmp3kBftC1ZlA,9
+semantio-0.0.3.dist-info/RECORD,,

semantio/tools/web_browser.py

@@ -1,153 +0,0 @@
-from typing import Dict, Any, Optional, List
-from playwright.async_api import async_playwright
-import asyncio
-import logging
-
-logger = logging.getLogger(__name__)
-
-class WebBrowserTool:
-    """
-    A tool for performing browser automation tasks using Playwright.
-    """
-
-    def __init__(self, headless: bool = True):
-        """
-        Initialize the WebBrowserTool.
-
-        Args:
-            headless (bool): Whether to run the browser in headless mode (default: True).
-        """
-        self.headless = headless
-        self.browser = None
-        self.context = None
-        self.page = None
-
-    async def start(self):
-        """
-        Start the browser and create a new context and page.
-        """
-        self.playwright = await async_playwright().start()
-        self.browser = await self.playwright.chromium.launch(headless=self.headless)
-        self.context = await self.browser.new_context()
-        self.page = await self.context.new_page()
-        logger.info("Browser started successfully.")
-
-    async def close(self):
-        """
-        Close the browser and cleanup resources.
-        """
-        if self.browser:
-            await self.browser.close()
-        await self.playwright.stop()
-        logger.info("Browser closed successfully.")
-
-    async def navigate(self, url: str) -> str:
-        """
-        Navigate to a specific URL.
-
-        Args:
-            url (str): The URL to navigate to.
-
-        Returns:
-            str: The page title after navigation.
-        """
-        if not self.page:
-            raise RuntimeError("Browser is not started. Call start() first.")
-
-        await self.page.goto(url)
-        title = await self.page.title()
-        logger.info(f"Navigated to {url}. Page title: {title}")
-        return title
-
-    async def fill_form(self, fields: Dict[str, str]) -> str:
-        """
-        Fill a form with the provided fields.
-
-        Args:
-            fields (Dict[str, str]): A dictionary of field names and values to fill.
-
-        Returns:
-            str: A success message.
-        """
-        if not self.page:
-            raise RuntimeError("Browser is not started. Call start() first.")
-
-        for field, value in fields.items():
-            await self.page.fill(f'input[name="{field}"]', value)
-            logger.info(f"Filled field '{field}' with value '{value}'.")
-
-        return "Form filled successfully."
-
-    async def click(self, selector: str) -> str:
-        """
-        Click an element on the page.
-
-        Args:
-            selector (str): The CSS selector of the element to click.
-
-        Returns:
-            str: A success message.
-        """
-        if not self.page:
-            raise RuntimeError("Browser is not started. Call start() first.")
-
-        await self.page.click(selector)
-        logger.info(f"Clicked element with selector '{selector}'.")
-        return f"Clicked element: {selector}"
-
-    async def scrape(self, selector: str) -> List[Dict[str, str]]:
-        """
-        Scrape data from the page.
-
-        Args:
-            selector (str): The CSS selector of the elements to scrape.
-
-        Returns:
-            List[Dict[str, str]]: A list of dictionaries containing the scraped data.
-        """
-        if not self.page:
-            raise RuntimeError("Browser is not started. Call start() first.")
-
-        elements = await self.page.query_selector_all(selector)
-        scraped_data = []
-        for element in elements:
-            text = await element.inner_text()
-            scraped_data.append({"text": text.strip()})
-            logger.info(f"Scraped text: {text.strip()}")
-
-        return scraped_data
-
-    async def execute_step(self, step: Dict[str, Any]) -> str:
-        """
-        Execute a browser automation step.
-
-        Args:
-            step (Dict[str, Any]): A dictionary containing the step details.
-                - "action": The action to perform (e.g., "navigate", "fill_form", "click", "scrape").
-                - "details": The details required for the action (e.g., URL, form fields, selector).
-                - "website": The website to perform the action on (optional).
-
-        Returns:
-            str: The result of the step execution.
-        """
-        action = step.get("action")
-        details = step.get("details")
-        website = step.get("website", "https://www.google.com")
-
-        if not self.page:
-            await self.start()
-
-        try:
-            if action == "navigate":
-                return await self.navigate(details)
-            elif action == "fill_form":
-                return await self.fill_form(details)
-            elif action == "click":
-                return await self.click(details)
-            elif action == "scrape":
-                return str(await self.scrape(details))
-            else:
-                return f"Unknown action: {action}"
-        except Exception as e:
-            logger.error(f"Error executing step: {e}")
-            return f"Error executing step: {e}"