aiagents4pharma 1.9.0__py3-none-any.whl → 1.15.0__py3-none-any.whl

aiagents4pharma/talk2competitors/agents/main_agent.py

@@ -0,0 +1,130 @@
+#!/usr/bin/env python3
+
+"""
+Main agent for the talk2competitors app.
+"""
+
+import logging
+from typing import Literal
+from dotenv import load_dotenv
+from langchain_core.language_models.chat_models import BaseChatModel
+from langchain_core.messages import AIMessage
+from langchain_openai import ChatOpenAI
+from langgraph.checkpoint.memory import MemorySaver
+from langgraph.graph import END, START, StateGraph
+from langgraph.types import Command
+from ..agents import s2_agent
+from ..config.config import config
+from ..state.state_talk2competitors import Talk2Competitors
+
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+load_dotenv()
+
+def make_supervisor_node(llm: BaseChatModel) -> str:
+    """
+    Creates a supervisor node following LangGraph patterns.
+
+    Args:
+        llm (BaseChatModel): The language model to use for generating responses.
+
+    Returns:
+        str: The supervisor node function.
+    """
+    # options = ["FINISH", "s2_agent"]
+
+    def supervisor_node(state: Talk2Competitors) -> Command[Literal["s2_agent", "__end__"]]:
+        """
+        Supervisor node that routes to appropriate sub-agents.
+
+        Args:
+            state (Talk2Competitors): The current state of the conversation.
+
+        Returns:
+            Command[Literal["s2_agent", "__end__"]]: The command to execute next.
+        """
+        logger.info("Supervisor node called")
+
+        messages = [{"role": "system", "content": config.MAIN_AGENT_PROMPT}] + state[
+            "messages"
+        ]
+        response = llm.invoke(messages)
+        goto = (
+            "FINISH"
+            if not any(
+                kw in state["messages"][-1].content.lower()
+                for kw in ["search", "paper", "find"]
+            )
+            else "s2_agent"
+        )
+
+        if goto == "FINISH":
+            return Command(
+                goto=END,
+                update={
+                    "messages": state["messages"]
+                    + [AIMessage(content=response.content)],
+                    "is_last_step": True,
+                    "current_agent": None,
+                },
+            )
+
+        return Command(
+            goto="s2_agent",
+            update={
+                "messages": state["messages"],
+                "is_last_step": False,
+                "current_agent": "s2_agent",
+            },
+        )
+
+    return supervisor_node
+
+def get_app(thread_id: str, llm_model ='gpt-4o-mini') -> StateGraph:
+    """
+    Returns the langraph app with hierarchical structure.
+
+    Args:
+        thread_id (str): The thread ID for the conversation.
+
+    Returns:
+        The compiled langraph app.
+    """
+    def call_s2_agent(state: Talk2Competitors) -> Command[Literal["__end__"]]:
+        """
+        Node for calling the S2 agent.
+
+        Args:
+            state (Talk2Competitors): The current state of the conversation.
+
+        Returns:
+            Command[Literal["__end__"]]: The command to execute next.
+        """
+        logger.info("Calling S2 agent")
+        app = s2_agent.get_app(thread_id, llm_model)
+        response = app.invoke(state)
+        logger.info("S2 agent completed")
+        return Command(
+            goto=END,
+            update={
+                "messages": response["messages"],
+                "papers": response.get("papers", []),
+                "is_last_step": True,
+                "current_agent": "s2_agent",
+            },
+        )
+    llm = ChatOpenAI(model=llm_model, temperature=0)
+    workflow = StateGraph(Talk2Competitors)
+
+    supervisor = make_supervisor_node(llm)
+    workflow.add_node("supervisor", supervisor)
+    workflow.add_node("s2_agent", call_s2_agent)
+
+    # Define edges
+    workflow.add_edge(START, "supervisor")
+    workflow.add_edge("s2_agent", END)
+
+    app = workflow.compile(checkpointer=MemorySaver())
+    logger.info("Main agent workflow compiled")
+    return app

aiagents4pharma/talk2competitors/agents/s2_agent.py

@@ -0,0 +1,75 @@
+#/usr/bin/env python3
+
+'''
+Agent for interacting with Semantic Scholar
+'''
+
+import logging
+from dotenv import load_dotenv
+from langchain_openai import ChatOpenAI
+from langgraph.graph import START, StateGraph
+from langgraph.prebuilt import create_react_agent
+from langgraph.checkpoint.memory import MemorySaver
+from ..config.config import config
+from ..state.state_talk2competitors import Talk2Competitors
+# from ..tools.s2 import s2_tools
+from ..tools.s2.search import search_tool
+from ..tools.s2.display_results import display_results
+from ..tools.s2.single_paper_rec import get_single_paper_recommendations
+from ..tools.s2.multi_paper_rec import get_multi_paper_recommendations
+
+load_dotenv()
+
+# Initialize logger
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+def get_app(uniq_id, llm_model='gpt-4o-mini'):
+    '''
+    This function returns the langraph app.
+    '''
+    def agent_s2_node(state: Talk2Competitors):
+        '''
+        This function calls the model.
+        '''
+        logger.log(logging.INFO, "Creating Agent_S2 node with thread_id %s", uniq_id)
+        response = model.invoke(state, {"configurable": {"thread_id": uniq_id}})
+        return response
+
+    # Define the tools
+    tools = [search_tool,
+            display_results,
+            get_single_paper_recommendations,
+            get_multi_paper_recommendations]
+
+    # Create the LLM
+    llm = ChatOpenAI(model=llm_model, temperature=0)
+    model = create_react_agent(
+                            llm,
+                            tools=tools,
+                            state_schema=Talk2Competitors,
+                            state_modifier=config.S2_AGENT_PROMPT,
+                            checkpointer=MemorySaver()
+                        )
+
+    # Define a new graph
+    workflow = StateGraph(Talk2Competitors)
+
+    # Define the two nodes we will cycle between
+    workflow.add_node("agent_s2", agent_s2_node)
+
+    # Set the entrypoint as `agent`
+    # This means that this node is the first one called
+    workflow.add_edge(START, "agent_s2")
+
+    # Initialize memory to persist state between graph runs
+    checkpointer = MemorySaver()
+
+    # Finally, we compile it!
+    # This compiles it into a LangChain Runnable,
+    # meaning you can use it as you would any other runnable.
+    # Note that we're (optionally) passing the memory when compiling the graph
+    app = workflow.compile(checkpointer=checkpointer)
+    logger.log(logging.INFO, "Compiled the graph")
+
+    return app

aiagents4pharma/talk2competitors/config/__init__.py

@@ -0,0 +1,5 @@
+"""
+This package contains configuration settings and prompts used by various AI agents
+"""
+
+from . import config

aiagents4pharma/talk2competitors/config/config.py

@@ -0,0 +1,110 @@
+"""Configuration module for AI agents handling paper searches and recommendations."""
+
+
+# pylint: disable=R0903
+class Config:
+    """Configuration class containing prompts for AI agents.
+
+    This class stores prompt templates used by various AI agents in the system,
+    particularly for academic paper searches and recommendations.
+    """
+
+    MAIN_AGENT_PROMPT = (
+        "You are a supervisory AI agent that routes user queries to specialized tools.\n"
+        "Your task is to select the most appropriate tool based on the user's request.\n\n"
+        "Available tools and their capabilities:\n\n"
+        "1. semantic_scholar_agent:\n"
+        "   - Search for academic papers and research\n"
+        "   - Get paper recommendations\n"
+        "   - Find similar papers\n"
+        "   USE FOR: Any queries about finding papers, academic research, "
+        "or getting paper recommendations\n\n"
+        "ROUTING GUIDELINES:\n\n"
+        "ALWAYS route to semantic_scholar_agent for:\n"
+        "- Finding academic papers\n"
+        "- Searching research topics\n"
+        "- Getting paper recommendations\n"
+        "- Finding similar papers\n"
+        "- Any query about academic literature\n\n"
+        "Approach:\n"
+        "1. Identify the core need in the user's query\n"
+        "2. Select the most appropriate tool based on the guidelines above\n"
+        "3. If unclear, ask for clarification\n"
+        "4. For multi-step tasks, focus on the immediate next step\n\n"
+        "Remember:\n"
+        "- Be decisive in your tool selection\n"
+        "- Focus on the immediate task\n"
+        "- Default to semantic_scholar_agent for any paper-finding tasks\n"
+        "- Ask for clarification if the request is ambiguous\n\n"
+        "When presenting paper search results, always use this exact format:\n\n"
+        "Remember to:\n"
+        "- Always remember to add the url\n"
+        "- Put URLs on the title line itself as markdown\n"
+        "- Maintain consistent spacing and formatting"
+    )
+
+    S2_AGENT_PROMPT = (
+        "You are a specialized academic research assistant with access to the following tools:\n\n"
+        "1. search_papers:\n"
+        "   USE FOR: General paper searches\n"
+        "   - Enhances search terms automatically\n"
+        "   - Adds relevant academic keywords\n"
+        "   - Focuses on recent research when appropriate\n\n"
+        "2. get_single_paper_recommendations:\n"
+        "   USE FOR: Finding papers similar to a specific paper\n"
+        "   - Takes a single paper ID\n"
+        "   - Returns related papers\n\n"
+        "3. get_multi_paper_recommendations:\n"
+        "   USE FOR: Finding papers similar to multiple papers\n"
+        "   - Takes multiple paper IDs\n"
+        "   - Finds papers related to all inputs\n\n"
+        "GUIDELINES:\n\n"
+        "For paper searches:\n"
+        "- Enhance search terms with academic language\n"
+        "- Include field-specific terminology\n"
+        '- Add "recent" or "latest" when appropriate\n'
+        "- Keep queries focused and relevant\n\n"
+        "For paper recommendations:\n"
+        "- Identify paper IDs (40-character hexadecimal strings)\n"
+        "- Use single_paper_recommendations for one ID\n"
+        "- Use multi_paper_recommendations for multiple IDs\n\n"
+        "Best practices:\n"
+        "1. Start with a broad search if no paper IDs are provided\n"
+        "2. Look for paper IDs in user input\n"
+        "3. Enhance search terms for better results\n"
+        "4. Consider the academic context\n"
+        "5. Be prepared to refine searches based on feedback\n\n"
+        "Remember:\n"
+        "- Always select the most appropriate tool\n"
+        "- Enhance search queries naturally\n"
+        "- Consider academic context\n"
+        "- Focus on delivering relevant results\n\n"
+        "IMPORTANT GUIDELINES FOR PAPER RECOMMENDATIONS:\n\n"
+        "For Multiple Papers:\n"
+        "- When getting recommendations for multiple papers, always use "
+        "get_multi_paper_recommendations tool\n"
+        "- DO NOT call get_single_paper_recommendations multiple times\n"
+        "- Always pass all paper IDs in a single call to get_multi_paper_recommendations\n"
+        '- Use for queries like "find papers related to both/all papers" or '
+        '"find similar papers to these papers"\n\n'
+        "For Single Paper:\n"
+        "- Use get_single_paper_recommendations when focusing on one specific paper\n"
+        "- Pass only one paper ID at a time\n"
+        '- Use for queries like "find papers similar to this paper" or '
+        '"get recommendations for paper X"\n'
+        "- Do not use for multiple papers\n\n"
+        "Examples:\n"
+        '- For "find related papers for both papers":\n'
+        "  ✓ Use get_multi_paper_recommendations with both paper IDs\n"
+        "  × Don't make multiple calls to get_single_paper_recommendations\n\n"
+        '- For "find papers related to the first paper":\n'
+        "  ✓ Use get_single_paper_recommendations with just that paper's ID\n"
+        "  × Don't use get_multi_paper_recommendations\n\n"
+        "Remember:\n"
+        "- Be precise in identifying which paper ID to use for single recommendations\n"
+        "- Don't reuse previous paper IDs unless specifically requested\n"
+        "- For fresh paper recommendations, always use the original paper ID"
+    )
+
+
+config = Config()

aiagents4pharma/talk2competitors/state/__init__.py

@@ -0,0 +1,5 @@
+'''
+This file is used to import all the modules in the package.
+'''
+
+from . import state_talk2competitors

aiagents4pharma/talk2competitors/state/state_talk2competitors.py

@@ -0,0 +1,32 @@
+"""
+This is the state file for the talk2comp agent.
+"""
+
+import logging
+from typing import Annotated, Any, Dict, Optional
+
+from langgraph.prebuilt.chat_agent_executor import AgentState
+from typing_extensions import NotRequired, Required
+
+# Configure logging
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+
+def replace_dict(existing: Dict[str, Any], new: Dict[str, Any]) -> Dict[str, Any]:
+    """Replace the existing dict with the new one."""
+    logger.info("Updating existing state %s with the state dict: %s", existing, new)
+    return new
+
+
+class Talk2Competitors(AgentState):
+    """
+    The state for the talk2comp agent, inheriting from AgentState.
+    """
+
+    papers: Annotated[Dict[str, Any], replace_dict]  # Changed from List to Dict
+    search_table: NotRequired[str]
+    next: str  # Required for routing in LangGraph
+    current_agent: NotRequired[Optional[str]]
+    is_last_step: Required[bool]  # Required field for LangGraph
+    llm_model: str

aiagents4pharma/talk2competitors/tests/__init__.py

@@ -0,0 +1,3 @@
+"""
+This module contains the test cases.
+"""

aiagents4pharma/talk2competitors/tests/test_langgraph.py

@@ -0,0 +1,274 @@
+"""
+Unit and integration tests for Talk2Competitors system.
+Each test focuses on a single, specific functionality.
+Tests are deterministic and independent of each other.
+"""
+
+from unittest.mock import Mock, patch
+
+import pytest
+from langchain_core.messages import AIMessage, HumanMessage
+
+from ..agents.main_agent import get_app, make_supervisor_node
+from ..state.state_talk2competitors import replace_dict
+from ..tools.s2.display_results import display_results
+from ..tools.s2.multi_paper_rec import get_multi_paper_recommendations
+from ..tools.s2.search import search_tool
+from ..tools.s2.single_paper_rec import get_single_paper_recommendations
+
+# pylint: disable=redefined-outer-name
+
+# Fixed test data for deterministic results
+MOCK_SEARCH_RESPONSE = {
+    "data": [
+        {
+            "paperId": "123",
+            "title": "Machine Learning Basics",
+            "abstract": "An introduction to ML",
+            "year": 2023,
+            "citationCount": 100,
+            "url": "https://example.com/paper1",
+            "authors": [{"name": "Test Author"}],
+        }
+    ]
+}
+
+MOCK_STATE_PAPER = {
+    "123": {
+        "Title": "Machine Learning Basics",
+        "Abstract": "An introduction to ML",
+        "Year": 2023,
+        "Citation Count": 100,
+        "URL": "https://example.com/paper1",
+    }
+}
+
+
+@pytest.fixture
+def initial_state():
+    """Create a base state for tests"""
+    return {
+        "messages": [],
+        "papers": {},
+        "is_last_step": False,
+        "current_agent": None,
+        "llm_model": "gpt-4o-mini",
+    }
+
+
+class TestMainAgent:
+    """Unit tests for main agent functionality"""
+
+    def test_supervisor_routes_search_to_s2(self, initial_state):
+        """Verifies that search-related queries are routed to S2 agent"""
+        llm_mock = Mock()
+        llm_mock.invoke.return_value = AIMessage(content="Search initiated")
+
+        supervisor = make_supervisor_node(llm_mock)
+        state = initial_state.copy()
+        state["messages"] = [HumanMessage(content="search for papers")]
+
+        result = supervisor(state)
+        assert result.goto == "s2_agent"
+        assert not result.update["is_last_step"]
+        assert result.update["current_agent"] == "s2_agent"
+
+    def test_supervisor_routes_general_to_end(self, initial_state):
+        """Verifies that non-search queries end the conversation"""
+        llm_mock = Mock()
+        llm_mock.invoke.return_value = AIMessage(content="General response")
+
+        supervisor = make_supervisor_node(llm_mock)
+        state = initial_state.copy()
+        state["messages"] = [HumanMessage(content="What is ML?")]
+
+        result = supervisor(state)
+        assert result.goto == "__end__"
+        assert result.update["is_last_step"]
+
+
+class TestS2Tools:
+    """Unit tests for individual S2 tools"""
+
+    def test_display_results_shows_papers(self, initial_state):
+        """Verifies display_results tool correctly returns papers from state"""
+        state = initial_state.copy()
+        state["papers"] = MOCK_STATE_PAPER
+        result = display_results.invoke(input={"state": state})
+        assert result == MOCK_STATE_PAPER
+        assert isinstance(result, dict)
+
+    @patch("requests.get")
+    def test_search_finds_papers(self, mock_get):
+        """Verifies search tool finds and formats papers correctly"""
+        mock_get.return_value.json.return_value = MOCK_SEARCH_RESPONSE
+        mock_get.return_value.status_code = 200
+
+        result = search_tool.invoke(
+            input={
+                "query": "machine learning",
+                "limit": 1,
+                "tool_call_id": "test123",
+                "id": "test123",
+            }
+        )
+
+        assert "papers" in result.update
+        assert "messages" in result.update
+        papers = result.update["papers"]
+        assert isinstance(papers, dict)
+        assert len(papers) > 0
+        paper = next(iter(papers.values()))
+        assert paper["Title"] == "Machine Learning Basics"
+        assert paper["Year"] == 2023
+
+    @patch("requests.get")
+    def test_single_paper_rec_basic(self, mock_get):
+        """Tests basic single paper recommendation functionality"""
+        mock_get.return_value.json.return_value = {
+            "recommendedPapers": [MOCK_SEARCH_RESPONSE["data"][0]]
+        }
+        mock_get.return_value.status_code = 200
+
+        result = get_single_paper_recommendations.invoke(
+            input={
+                "paper_id": "123",
+                "limit": 1,
+                "tool_call_id": "test123",
+                "id": "test123",
+            }
+        )
+        assert "papers" in result.update
+        assert len(result.update["messages"]) == 1
+
+    @patch("requests.get")
+    def test_single_paper_rec_with_optional_params(self, mock_get):
+        """Tests single paper recommendations with year parameter"""
+        mock_get.return_value.json.return_value = {
+            "recommendedPapers": [MOCK_SEARCH_RESPONSE["data"][0]]
+        }
+        mock_get.return_value.status_code = 200
+
+        result = get_single_paper_recommendations.invoke(
+            input={
+                "paper_id": "123",
+                "limit": 1,
+                "year": "2023-",
+                "tool_call_id": "test123",
+                "id": "test123",
+            }
+        )
+        assert "papers" in result.update
+
+    @patch("requests.post")
+    def test_multi_paper_rec_basic(self, mock_post):
+        """Tests basic multi-paper recommendation functionality"""
+        mock_post.return_value.json.return_value = {
+            "recommendedPapers": [MOCK_SEARCH_RESPONSE["data"][0]]
+        }
+        mock_post.return_value.status_code = 200
+
+        result = get_multi_paper_recommendations.invoke(
+            input={
+                "paper_ids": ["123", "456"],
+                "limit": 1,
+                "tool_call_id": "test123",
+                "id": "test123",
+            }
+        )
+        assert "papers" in result.update
+        assert len(result.update["messages"]) == 1
+
+    @patch("requests.post")
+    def test_multi_paper_rec_with_optional_params(self, mock_post):
+        """Tests multi-paper recommendations with all optional parameters"""
+        mock_post.return_value.json.return_value = {
+            "recommendedPapers": [MOCK_SEARCH_RESPONSE["data"][0]]
+        }
+        mock_post.return_value.status_code = 200
+
+        result = get_multi_paper_recommendations.invoke(
+            input={
+                "paper_ids": ["123", "456"],
+                "limit": 1,
+                "year": "2023-",
+                "tool_call_id": "test123",
+                "id": "test123",
+            }
+        )
+        assert "papers" in result.update
+        assert len(result.update["messages"]) == 1
+
+    @patch("requests.get")
+    def test_single_paper_rec_empty_response(self, mock_get):
+        """Tests single paper recommendations with empty response"""
+        mock_get.return_value.json.return_value = {"recommendedPapers": []}
+        mock_get.return_value.status_code = 200
+
+        result = get_single_paper_recommendations.invoke(
+            input={
+                "paper_id": "123",
+                "limit": 1,
+                "tool_call_id": "test123",
+                "id": "test123",
+            }
+        )
+        assert "papers" in result.update
+        assert len(result.update["papers"]) == 0
+
+    @patch("requests.post")
+    def test_multi_paper_rec_empty_response(self, mock_post):
+        """Tests multi-paper recommendations with empty response"""
+        mock_post.return_value.json.return_value = {"recommendedPapers": []}
+        mock_post.return_value.status_code = 200
+
+        result = get_multi_paper_recommendations.invoke(
+            input={
+                "paper_ids": ["123", "456"],
+                "limit": 1,
+                "tool_call_id": "test123",
+                "id": "test123",
+            }
+        )
+        assert "papers" in result.update
+        assert len(result.update["papers"]) == 0
+
+
+def test_state_replace_dict():
+    """Verifies state dictionary replacement works correctly"""
+    existing = {"key1": "value1", "key2": "value2"}
+    new = {"key3": "value3"}
+    result = replace_dict(existing, new)
+    assert result == new
+    assert isinstance(result, dict)
+
+
+@pytest.mark.integration
+def test_end_to_end_search_workflow(initial_state):
+    """Integration test: Complete search workflow"""
+    with (
+        patch("requests.get") as mock_get,
+        patch("langchain_openai.ChatOpenAI") as mock_llm,
+    ):
+        mock_get.return_value.json.return_value = MOCK_SEARCH_RESPONSE
+        mock_get.return_value.status_code = 200
+
+        llm_instance = Mock()
+        llm_instance.invoke.return_value = AIMessage(content="Search completed")
+        mock_llm.return_value = llm_instance
+
+        app = get_app("test_integration")
+        test_state = initial_state.copy()
+        test_state["messages"] = [HumanMessage(content="search for ML papers")]
+
+        config = {
+            "configurable": {
+                "thread_id": "test_integration",
+                "checkpoint_ns": "test",
+                "checkpoint_id": "test123",
+            }
+        }
+
+        response = app.invoke(test_state, config)
+        assert "papers" in response
+        assert len(response["messages"]) > 0

aiagents4pharma/talk2competitors/tools/__init__.py

@@ -0,0 +1,7 @@
+#!/usr/bin/env python3
+
+'''
+Import statements
+'''
+
+from . import s2

aiagents4pharma/talk2competitors/tools/s2/__init__.py

@@ -0,0 +1,8 @@
+'''
+This file is used to import all the modules in the package.
+'''
+
+from . import display_results
+from . import multi_paper_rec
+from . import search
+from . import single_paper_rec

aiagents4pharma/talk2competitors/tools/s2/display_results.py

@@ -0,0 +1,25 @@
+#!/usr/bin/env python3
+
+'''
+This tool is used to display the table of studies.
+'''
+
+import logging
+from typing import Annotated
+from langchain_core.tools import tool
+from langgraph.prebuilt import InjectedState
+
+# Configure logging
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+@tool('display_results')
+def display_results(state: Annotated[dict, InjectedState]):
+    """
+    Display the papers in the state.
+
+    Args:
+        state (dict): The state of the agent.
+    """
+    logger.info("Displaying papers from the state")
+    return state["papers"]