import os from PIL import Image import sys import pandas as pd import requests from omegaconf import OmegaConf import streamlit as st from streamlit_pills import pills from dotenv import load_dotenv load_dotenv(override=True) from pydantic import Field, BaseModel from vectara_agent.agent import Agent, AgentStatusType from vectara_agent.tools import ToolsFactory tickers = { "AAPL": "Apple Computer", "GOOG": "Google", "AMZN": "Amazon", "SNOW": "Snowflake", "TEAM": "Atlassian", "TSLA": "Tesla", "NVDA": "Nvidia", "MSFT": "Microsoft", "AMD": "Advanced Micro Devices", "INTC": "Intel", "NFLX": "Netflix", } years = [2020, 2021, 2022, 2023, 2024] initial_prompt = "How can I help you today?" def create_tools(cfg): def get_company_info() -> list[str]: """ Returns a dictionary of companies you can query about. Always check this before using any other tool. The output is a dictionary of valid ticker symbols mapped to company names. You can use this to identify the companies you can query about, and their ticker information. """ return tickers def get_valid_years() -> list[str]: """ Returns a list of the years for which financial reports are available. Always check this before using any other tool. """ return years # Tool to get the income statement for a given company and year using the FMP API def get_income_statement( ticker=Field(description="the ticker symbol of the company."), year=Field(description="the year for which to get the income statement."), ) -> str: """ Get the income statement for a given company and year using the FMP (https://financialmodelingprep.com) API. Returns a dictionary with the income statement data. All data is in USD, but you can convert it to more compact form like K, M, B. """ fmp_api_key = os.environ.get("FMP_API_KEY", None) if fmp_api_key is None: return "FMP_API_KEY environment variable not set. This tool does not work." url = f"https://financialmodelingprep.com/api/v3/income-statement/{ticker}?apikey={fmp_api_key}" response = requests.get(url) if response.status_code == 200: data = response.json() income_statement = pd.DataFrame(data) income_statement["date"] = pd.to_datetime(income_statement["date"]) income_statement_specific_year = income_statement[ income_statement["date"].dt.year == int(year) ] values_dict = income_statement_specific_year.to_dict(orient="records")[0] return f"Financial results: {', '.join([f'{key}: {value}' for key, value in values_dict.items() if key not in ['date', 'cik', 'link', 'finalLink']])}" else: return "FMP API returned error. This tool does not work." class QueryTranscriptsArgs(BaseModel): query: str = Field(..., description="The user query.") year: int = Field(..., description=f"The year. An integer between {min(years)} and {max(years)}.") ticker: str = Field(..., description=f"The company ticker. Must be a valid ticket symbol from the list {tickers.keys()}.") tools_factory = ToolsFactory(vectara_api_key=cfg.api_key, vectara_customer_id=cfg.customer_id, vectara_corpus_id=cfg.corpus_id) ask_transcripts = tools_factory.create_rag_tool( tool_name = "ask_transcripts", tool_description = """ Given a company name and year, responds to a user question about the company, based on analyst call transcripts about the company's financial reports for that year. You can ask this tool any question about the compaany including risks, opportunities, financial performance, competitors and more. """, tool_args_schema = QueryTranscriptsArgs, reranker = "multilingual_reranker_v1", rerank_k = 100, n_sentences_before = 2, n_sentences_after = 2, lambda_val = 0.005, summary_num_results = 10, vectara_summarizer = 'vectara-summary-ext-24-05-med-omni', include_citations = False, ) return (tools_factory.get_tools( [ get_company_info, get_valid_years, get_income_statement, ] ) + tools_factory.standard_tools() + tools_factory.financial_tools() + tools_factory.guardrail_tools() + [ask_transcripts] ) def initialize_agent(_cfg): if 'agent' in st.session_state: return st.session_state.agent financial_bot_instructions = """ - You are a helpful financial assistant, with expertise in financial reporting, in conversation with a user. - Never discuss politics, and always respond politely. - Respond in a compact format by using appropriate units of measure (e.g., K for thousands, M for millions, B for billions). Do not report the same number twice (e.g. $100K and 100,000 USD). - Always check the get_company_info and get_valid_years tools to validate company and year are valid. - When querying a tool for a numeric value or KPI, use a concise and non-ambiguous description of what you are looking for. - If you calculate a metric, make sure you have all the necessary information to complete the calculation. Don't guess. """ def update_func(status_type: AgentStatusType, msg: str): if status_type != AgentStatusType.AGENT_UPDATE: output = f"{status_type.value} - {msg}" st.session_state.log_messages.append(output) agent = Agent( tools=create_tools(_cfg), topic="Financial data, annual reports and 10-K filings", custom_instructions=financial_bot_instructions, update_func=update_func ) agent.report() return agent def toggle_logs(): st.session_state.show_logs = not st.session_state.show_logs def show_example_questions(): if len(st.session_state.example_messages) > 0 and st.session_state.first_turn: selected_example = pills("Queries to Try:", st.session_state.example_messages, index=None) if selected_example: st.session_state.ex_prompt = selected_example st.session_state.first_turn = False return True return False def launch_bot(): def reset(): st.session_state.messages = [{"role": "assistant", "content": initial_prompt, "avatar": "🦖"}] st.session_state.thinking_message = "Agent at work..." st.session_state.log_messages = [] st.session_state.prompt = None st.session_state.ex_prompt = None st.session_state.show_logs = False st.session_state.first_turn = True st.set_page_config(page_title="Financial Assistant", layout="wide") if 'cfg' not in st.session_state: cfg = OmegaConf.create({ 'customer_id': str(os.environ['VECTARA_CUSTOMER_ID']), 'corpus_id': str(os.environ['VECTARA_CORPUS_ID']), 'api_key': str(os.environ['VECTARA_API_KEY']), 'examples': os.environ.get('QUERY_EXAMPLES', None) }) st.session_state.cfg = cfg st.session_state.ex_prompt = None example_messages = [example.strip() for example in cfg.examples.split(",")] if cfg.examples else [] st.session_state.example_messages = [em for em in example_messages if len(em)>0] reset() cfg = st.session_state.cfg if 'agent' not in st.session_state: st.session_state.agent = initialize_agent(cfg) # left side content with st.sidebar: image = Image.open('Vectara-logo.png') st.image(image, width=175) st.markdown("## Welcome to the financial assistant demo.\n\n\n") companies = ", ".join(tickers.values()) st.markdown( f"This assistant can help you with any questions about the financials of several companies:\n\n **{companies}**.\n" ) st.markdown("\n\n") bc1, _ = st.columns([1, 1]) with bc1: if st.button('Start Over'): reset() st.rerun() st.markdown("---") st.markdown( "## How this works?\n" "This app was built with [Vectara](https://vectara.com).\n\n" "It demonstrates the use of Agentic RAG functionality with Vectara" ) st.markdown("---") if "messages" not in st.session_state.keys(): reset() # Display chat messages for message in st.session_state.messages: with st.chat_message(message["role"], avatar=message["avatar"]): st.write(message["content"]) example_container = st.empty() with example_container: if show_example_questions(): example_container.empty() st.session_state.first_turn = False st.rerun() # User-provided prompt if st.session_state.ex_prompt: prompt = st.session_state.ex_prompt else: prompt = st.chat_input() if prompt: st.session_state.messages.append({"role": "user", "content": prompt, "avatar": '🧑‍💻'}) st.session_state.prompt = prompt # Save the prompt in session state st.session_state.log_messages = [] st.session_state.show_logs = False with st.chat_message("user", avatar='🧑‍💻'): print(f"Starting new question: {prompt}\n") st.write(prompt) st.session_state.ex_prompt = None # Generate a new response if last message is not from assistant if st.session_state.prompt: with st.chat_message("assistant", avatar='🤖'): with st.spinner(st.session_state.thinking_message): res = st.session_state.agent.chat(st.session_state.prompt) res = res.replace('$', '\\$') # escape dollar sign for markdown message = {"role": "assistant", "content": res, "avatar": '🤖'} st.session_state.messages.append(message) st.markdown(res) st.session_state.ex_prompt = None st.session_state.prompt = None st.rerun() log_placeholder = st.empty() with log_placeholder.container(): if st.session_state.show_logs: st.button("Hide Logs", on_click=toggle_logs) for msg in st.session_state.log_messages: st.text(msg) else: if len(st.session_state.log_messages) > 0: st.button("Show Logs", on_click=toggle_logs) sys.stdout.flush() if __name__ == "__main__": launch_bot()