Tool calling apps (#8)

Co-authored-by: Sam Julien <sam.julien@writer.com>

Author: yanomaly

finance-dashboard/charts.py

@@ -72,11 +72,11 @@ def update_scatter_chart(state):
     )
 
     # Set axis titles
-    fig.update_yaxes(title_text=f"{state["symbol"]} Stock Price", secondary_y=False)
+    fig.update_yaxes(title_text=f"{state['symbol']} Stock Price", secondary_y=False)
     fig.update_yaxes(title_text="S&P 500", secondary_y=True)
 
     # Update layout
-    fig.update_layout(height=550, title_text=f"{state["symbol"]} Stock vs the S&P 500", title_x = 0.5, title_y = 0.9, legend=dict(
+    fig.update_layout(height=550, title_text=f"{state['symbol']} Stock vs the S&P 500", title_x = 0.5, title_y = 0.9, legend=dict(
             orientation='h',
             yanchor='top',
             y=-0.2,  # Adjust this value as needed

finance-dashboard/main.py

@@ -3,7 +3,7 @@
 import pandas as pd
 from prompts import stock_prompts, income_prompts, earnings_prompt
 from stock_data import download_data, download_sp500, stock_news, _one_day_data, income_statement, earnings_calls
-from charts import update_scatter_chart
+from charts import update_scatter_chart, handle_click
 from dotenv import load_dotenv
 import os
 
@@ -34,12 +34,12 @@ def _refresh_window(state):
 # Summarize earnings call using Palmyra-Fin model
 def summarize_earnings(state):
     _refresh_window(state)  
-    state["message"] = f"% {state["symbol"]} earnings call will be summarized here"
+    state["message"] = f"% {state['symbol']} earnings call will be summarized here"
 
     earnings_transcript = state["earnings_transcript"]
     prompt = earnings_prompt.format(earnings_transcript=earnings_transcript)
     submission = writer.ai.complete(prompt, config={"model": "palmyra-fin-32k", "temperature": 0.7, "max_tokens": 8192})
-    state["message"] = f"+ {state["symbol"]} earnings call summary"
+    state["message"] = f"+ {state['symbol']} earnings call summary"
     state["analysis"] = submission.strip()
     state["show_analysis_text"]["visible"] = True
 
@@ -53,10 +53,10 @@ def prompt_parameters_lang(state,payload):
 
 def generate_stock_analysis(state):
     _refresh_window(state)  
-    if(state["prompt_parameters_lang"] == ""):
+    if state["prompt_parameters_lang"] == "":
         state["prompt_parameters_lang"] == "English"
 
-    state["message"] = f"% {state["symbol"]} trends will be analyzed here in {state['prompt_parameters_lang']}"
+    state["message"] = f"% {state['symbol']} trends will be analyzed here in {state['prompt_parameters_lang']}"
     stock_name = state["symbol"]
     stock_data = state["main_df"][:365]
 
@@ -68,7 +68,7 @@ def generate_stock_analysis(state):
     prompt = stock_prompts.format(language=language, stock_name=stock_name,words=integer_value,stock_data=stock_data)
     submission = writer.ai.complete(prompt, config={"model": "palmyra-fin-32k", "temperature": 0.7, "max_tokens": 8192})
     state["analysis"] = submission.strip()
-    state["message"] = f"+ {state["symbol"]} trends analyzed"
+    state["message"] = f"+ {state['symbol']} trends analyzed"
     state["show_analysis_text"]["visible"] = True
     state["show_analysis_text"]["language"] = True
 
@@ -77,7 +77,7 @@ def generate_stock_analysis(state):
 
 def generate_income_analysis(state):
     _refresh_window(state)
-    state["message"] = f"% {state["symbol"]} income statement will be visualized here"
+    state["message"] = f"% {state['symbol']} income statement will be visualized here"
     stock_name = state["symbol"]
     stock_data = state["main_df"][:365]
     income_statement_data = state["income_statement_df"][:365]
@@ -88,7 +88,7 @@ def generate_income_analysis(state):
     )
     submission = writer.ai.complete(prompt, config={"model": "palmyra-fin-32k", "temperature": 0.7, "max_tokens": 8192})
     state["analysis"] = submission.strip()
-    state["message"] = f"+ {state["symbol"]} income statement visualized"
+    state["message"] = f"+ {state['symbol']} income statement visualized"
     state["show_income_metrics"]["visible"] = True
     state["show_analysis_text"]["visible"] = True
     state["show_bar_graph"]["visible"] = True
@@ -107,7 +107,6 @@ def _get_main_df(filename):
         "last_24_hours_open": "168.76",
         "last_24_hours_high": "169.72",
         "last_24_hours_low": "167.50",
-        "message": None,
         "main_df": _get_main_df("daily_IBM.csv"),
         "main_df_subset": _get_main_df("daily_IBM.csv"),
         "symbol": "AAPL",

financial-tools-chat/.env.example

@@ -0,0 +1,2 @@
+WRITER_API_KEY=your_writer_api_key
+EARNINGS_ANALYSIS_APP_ID=your_zero_code_app_id

financial-tools-chat/.wf/components-page-0-c0f99a9e-5004-4e75-a6c6-36f17490b134.jsonl

@@ -0,0 +1,19 @@
+{"id": "c0f99a9e-5004-4e75-a6c6-36f17490b134", "type": "page", "content": {"pageMode": "compact"}, "handlers": {}, "isCodeManaged": false, "parentId": "root", "position": 0}
+{"id": "4kukoixkdb66ixah", "type": "header", "content": {"text": "Financial tools chat"}, "handlers": {}, "isCodeManaged": false, "parentId": "c0f99a9e-5004-4e75-a6c6-36f17490b134", "position": 0}
+{"id": "kldz702eixpb8xnq", "type": "columns", "content": {"cssClasses": "column-container"}, "handlers": {}, "isCodeManaged": false, "parentId": "c0f99a9e-5004-4e75-a6c6-36f17490b134", "position": 1}
+{"id": "4nklk9e1wqetdhln", "type": "column", "content": {"width": "45"}, "handlers": {}, "isCodeManaged": false, "parentId": "kldz702eixpb8xnq", "position": 0}
+{"id": "pi01zfrzcoxpiqxu", "type": "chatbot", "content": {"avatarBackgroundColor": "#000000", "buttonColor": "#000000", "conversation": "@{conversation}", "cssClasses": "chat", "useMarkdown": "yes"}, "handlers": {"wf-chatbot-message": "message_handler"}, "isCodeManaged": false, "parentId": "4nklk9e1wqetdhln", "position": 0, "visible": {"binding": "", "expression": true, "reversed": false}}
+{"id": "52rbj5jd79undjij", "type": "column", "content": {"width": "50"}, "handlers": {}, "isCodeManaged": false, "parentId": "kldz702eixpb8xnq", "position": 1, "visible": {"binding": "visual_block_visible", "expression": "custom", "reversed": false}}
+{"id": "e8iafahde157xsgr", "type": "tabs", "content": {"cssClasses": "visualization"}, "handlers": {}, "isCodeManaged": false, "parentId": "52rbj5jd79undjij", "position": 0}
+{"id": "stock", "type": "tab", "content": {"cssClasses": "", "name": "Stock"}, "handlers": {}, "isCodeManaged": false, "parentId": "e8iafahde157xsgr", "position": 0, "visible": {"binding": "", "expression": "custom", "reversed": false}}
+{"id": "er0t7muikaatylvt", "type": "plotlygraph", "content": {"cssClasses": "stock", "spec": "@{stock_chart}"}, "handlers": {}, "isCodeManaged": false, "parentId": "stock", "position": 0, "visible": {"binding": "", "expression": true, "reversed": false}}
+{"id": "4n2cykskypjd6qo4", "type": "text", "content": {"text": "@{stock_analysis}", "useMarkdown": "yes"}, "handlers": {}, "isCodeManaged": false, "parentId": "stock", "position": 1}
+{"id": "income", "type": "tab", "content": {"cssClasses": "", "name": "Income"}, "handlers": {}, "isCodeManaged": false, "parentId": "e8iafahde157xsgr", "position": 1, "visible": {"binding": "", "expression": "custom", "reversed": false}}
+{"id": "rqbqheyr0o5hs4fj", "type": "plotlygraph", "content": {"cssClasses": "income", "spec": "@{income_chart}"}, "handlers": {}, "isCodeManaged": false, "parentId": "income", "position": 0, "visible": {"binding": "", "expression": true, "reversed": false}}
+{"id": "ssw8zcojyuup2aqn", "type": "text", "content": {"text": "@{income_analysis}", "useMarkdown": "yes"}, "handlers": {}, "isCodeManaged": false, "parentId": "income", "position": 1, "visible": {"binding": "", "expression": "custom", "reversed": false}}
+{"id": "earnings", "type": "tab", "content": {"cssClasses": "", "name": "Earnings"}, "handlers": {}, "isCodeManaged": false, "parentId": "e8iafahde157xsgr", "position": 2, "visible": {"binding": "", "expression": "custom", "reversed": false}}
+{"id": "3um69zhvgxb6efbh", "type": "text", "content": {"text": "@{earnings_analysis}", "useMarkdown": "yes"}, "handlers": {}, "isCodeManaged": false, "parentId": "earnings", "position": 0}
+{"id": "qwoll83rzy43r3rr", "type": "column", "content": {"cssClasses": "buttons", "width": "1"}, "handlers": {}, "isCodeManaged": false, "parentId": "kldz702eixpb8xnq", "position": 2, "visible": {"binding": "", "expression": "custom", "reversed": false}}
+{"id": "19f02wo3koxaipwp", "type": "button", "content": {"buttonColor": "#ffffff", "buttonTextColor": "#000000", "cssClasses": "", "icon": "arrow_back_ios", "text": ""}, "handlers": {"wf-click": "open_visualization"}, "isCodeManaged": false, "parentId": "qwoll83rzy43r3rr", "position": 0, "visible": {"binding": "visual_block_visible", "expression": "custom", "reversed": true}}
+{"id": "72yjtz3kkrb5zr16", "type": "button", "content": {"buttonColor": "#ffffff", "buttonTextColor": "#241f31", "cssClasses": "", "icon": "arrow_forward_ios", "text": ""}, "handlers": {"wf-click": "close_visualization"}, "isCodeManaged": false, "parentId": "qwoll83rzy43r3rr", "position": 1, "visible": {"binding": "visual_block_visible", "expression": "custom", "reversed": false}}
+{"id": "kphyjxow4ppazmw1", "type": "button", "content": {"buttonColor": "#ffffff", "buttonTextColor": "#000000", "icon": "delete", "text": ""}, "handlers": {"wf-click": "clear_visualization"}, "isCodeManaged": false, "parentId": "qwoll83rzy43r3rr", "position": 2}

financial-tools-chat/.wf/components-root.jsonl

@@ -0,0 +1 @@
+{"id": "root", "type": "root", "content": {"appName": "Financial tools chat"}, "handlers": {}, "isCodeManaged": false, "position": 0}

financial-tools-chat/.wf/components-workflows_root.jsonl

@@ -0,0 +1 @@
+{"id": "workflows_root", "type": "workflows_root", "content": {}, "handlers": {}, "isCodeManaged": false, "position": 0, "visible": {"binding": "", "expression": true, "reversed": false}}

financial-tools-chat/.wf/metadata.json

@@ -0,0 +1,3 @@
+{
+    "writer_version": "0.8.1"
+}

financial-tools-chat/README.md

@@ -0,0 +1,3 @@
+This app was created using Writer Framework.
+
+To learn more about it, visit https://dev.writer.com/framework

financial-tools-chat/chat_tools.py

@@ -0,0 +1,185 @@
+import json
+import os
+from io import StringIO
+
+import pandas as pd
+import plotly.express as px
+import plotly.graph_objects as go
+import yfinance as yf
+from dotenv import load_dotenv
+from plotly.subplots import make_subplots
+from prompts import income_prompt, stock_prompt
+from writer.ai import apps, complete
+
+load_dotenv()
+
+
+def _get_stock_analysis(symbol: str) -> str:
+    config = {"model": "palmyra-fin-32k", "temperature": 0.7, "max_tokens": 8192}
+    data = _get_stock_data(symbol)
+    prompt = stock_prompt.format(name=symbol, data=data)
+    stock_analysis = complete(prompt, config=config)
+    return stock_analysis
+
+
+def _get_stock_data(symbol: str) -> str:
+    df = yf.download(symbol, period="5y", interval="5d")
+    df = df.reset_index()
+    df = df.sort_values(by="Date", ascending=False)
+    df = df.round({"Open": 2, "High": 2, "Low": 2, "Close": 2, "Adj Close": 2})
+    df["Date"] = pd.to_datetime(df["Date"])
+    return df.to_csv(index=False)
+
+
+def _get_income_analysis(symbol: str) -> str:
+    config = {"model": "palmyra-fin-32k", "temperature": 0.7, "max_tokens": 8192}
+    data = _get_income_data(symbol)
+    prompt = income_prompt.format(name=symbol, data=data)
+    income_analysis = complete(prompt, config=config)
+    return income_analysis
+
+
+def _get_income_data(symbol: str) -> str:
+    quarterly_income_stmt = yf.Ticker(symbol).quarterly_income_stmt
+    df = pd.DataFrame(quarterly_income_stmt)
+    df.columns = pd.to_datetime(df.columns).strftime("%Y-%m-%d")
+    return df.to_csv()
+
+
+def _get_earnings_analysis(symbol: str) -> str:
+    data = _get_earnings_data(symbol)
+    earnings_analysis = apps.generate_content(
+        application_id=os.getenv("EARNINGS_ANALYSIS_APP_ID", ""),
+        input_dict={
+            "name": symbol,
+            "data": data,
+        },
+    )
+    return earnings_analysis
+
+
+def _get_earnings_data(symbol: str) -> str:
+    with open("earnings-data.json", "r") as file:
+        earnings_transcript = json.load(file)
+    if earnings_transcript:
+        for item in earnings_transcript:
+            if item["symbol"] == symbol:
+                return item["content"]
+    else:
+        return "No earnings transcript found."
+
+
+def get_stock_chart(symbol: str) -> str:
+    symbol_io = StringIO(_get_stock_data(symbol))
+    sp500_io = StringIO(_get_stock_data("^GSPC"))
+
+    symbol_df = pd.read_csv(symbol_io)
+    sp500_df = pd.read_csv(sp500_io)
+
+    fig = make_subplots(specs=[[{"secondary_y": True}]])
+
+    fig.add_trace(
+        go.Scatter(x=symbol_df["Date"], y=symbol_df["Open"], name=symbol, mode="lines"),
+        secondary_y=False,
+    )
+
+    fig.add_trace(
+        go.Scatter(
+            x=sp500_df["Date"], y=sp500_df["Open"], name="S&P 500", mode="lines"
+        ),
+        secondary_y=True,
+    )
+
+    fig.update_yaxes(title_text=f"{symbol} Stock Price", secondary_y=False)
+    fig.update_yaxes(title_text="S&P 500", secondary_y=True)
+
+    fig.update_layout(
+        height=450,
+        title_text=f"{symbol} Stock vs the S&P 500",
+        title_x=0.5,
+        title_y=0.9,
+        legend=dict(
+            orientation="h",
+            yanchor="top",
+            y=-0.2,
+            xanchor="center",
+            x=0.5,
+        ),
+    )
+
+    return fig.to_json()
+
+
+def get_income_chart(symbol: str) -> str:
+    string_io = StringIO(_get_income_data(symbol))
+
+    df = pd.read_csv(string_io, index_col=0)
+    df_filtered = df.loc[["Total Revenue", "Net Income", "Operating Income"]]
+
+    df_transposed = df_filtered.transpose().reset_index()
+    df_transposed = df_transposed.melt(
+        id_vars=["index"], var_name="Metric", value_name="Value"
+    )
+
+    fig = px.bar(
+        df_transposed,
+        x="index",
+        y="Value",
+        color="Metric",
+        barmode="group",
+        labels={"index": "", "Value": ""},
+        title=f"Summary of Quarterly Income Statement for {symbol}",
+    )
+
+    fig.update_layout(
+        height=400,
+        legend=dict(orientation="h", yanchor="top", y=-0.2, xanchor="center", x=0.5),
+    )
+
+    return fig.to_json()
+
+
+stock_analysis_tool = {
+    "type": "function",
+    "name": "get_stock_analysis",
+    "callable": _get_stock_analysis,
+    "parameters": {
+        "symbol": {
+            "type": "string",
+            "description": "Symbol to compose stock data analysis.",
+        },
+    },
+}
+
+
+income_analysis_tool = {
+    "type": "function",
+    "name": "get_income_analysis",
+    "callable": _get_income_analysis,
+    "parameters": {
+        "symbol": {
+            "type": "string",
+            "description": "Symbol to compose income statement analysis.",
+        },
+    },
+}
+
+
+earnings_analysis_tool = {
+    "type": "function",
+    "name": "get_earnings_analysis",
+    "callable": _get_earnings_analysis,
+    "parameters": {
+        "symbol": {
+            "type": "string",
+            "description": "Symbol to fetch earnings data and compose analysis.",
+        },
+    },
+}
+
+
+tools = [
+    stock_analysis_tool,
+    income_analysis_tool,
+    earnings_analysis_tool,
+]