app.py

# --- Dependencies ---
from dotenv import load_dotenv
from openai import OpenAI
import json
import os
import requests
from pypdf import PdfReader
import gradio as gr

# Load environment variables from .env file (e.g. OPENAI_API_KEY, PUSHOVER_TOKEN)
load_dotenv(override=True)


# --- Pushover Integration (mobile notifications) ---
def push(text):
    """Send a notification to your phone via the Pushover API."""
    requests.post(
        "https://api.pushover.net/1/messages.json",
        data={
            "token": os.getenv("PUSHOVER_TOKEN"),
            "user": os.getenv("PUSHOVER_USER"),
            "message": text,
        }
    )


# --- Tool functions (callable by the AI when it decides to) ---
def record_user_details(email, name="Name not provided", notes="not provided"):
    """When a user wants to get in touch: send their contact info to your phone and acknowledge."""
    push(f"Recording {name} with email {email} and notes {notes}")
    return {"recorded": "ok"}


def record_unknown_question(question):
    """When the AI doesn't know the answer: log the question so you can follow up later."""
    push(f"Recording {question}")
    return {"recorded": "ok"}


# --- Tool schemas (OpenAI function calling format) ---
# These JSON objects describe each tool so the model knows when and how to call them.
record_user_details_json = {
    "name": "record_user_details",
    "description": "Use this tool to record that a user is interested in being in touch and provided an email address",
    "parameters": {
        "type": "object",
        "properties": {
            "email": {
                "type": "string",
                "description": "The email address of this user"
            },
            "name": {
                "type": "string",
                "description": "The user's name, if they provided it"
            }
            ,
            "notes": {
                "type": "string",
                "description": "Any additional information about the conversation that's worth recording to give context"
            }
        },
        "required": ["email"],
        "additionalProperties": False
    }
}

record_unknown_question_json = {
    "name": "record_unknown_question",
    "description": "Always use this tool to record any question that couldn't be answered as you didn't know the answer",
    "parameters": {
        "type": "object",
        "properties": {
            "question": {
                "type": "string",
                "description": "The question that couldn't be answered"
            },
        },
        "required": ["question"],
        "additionalProperties": False
    }
}

# List of tools exposed to the AI model (OpenAI function-calling format)
tools = [{"type": "function", "function": record_user_details_json},
        {"type": "function", "function": record_unknown_question_json}]


# --- Main agent: persona chatbot ---
class Me:

    def __init__(self):
        """Initialize the agent: connect to OpenAI and load persona knowledge from files."""
        self.openai = OpenAI()
        self.name = "Harold Malécot"
        # Load LinkedIn profile text from PDF (one string per page concatenated)
        reader = PdfReader("linkedin.pdf")
        self.linkedin = ""
        for page in reader.pages:
            text = page.extract_text()
            if text:
                self.linkedin += text
        # Use professional email instead of personal one when displayed/shared
        self.linkedin = self.linkedin.replace("harold.malecot@proton.me", "harold.job@proton.me")
        # Load additional summary text (e.g. bio, key points)
        with open("summary.txt", "r", encoding="utf-8") as f:
            self.summary = f.read()


    def handle_tool_call(self, tool_calls):
        """Run each tool the model requested and return formatted responses for the next API call."""
        results = []
        for tool_call in tool_calls:
            tool_name = tool_call.function.name
            arguments = json.loads(tool_call.function.arguments)
            print(f"Tool called: {tool_name}", flush=True)
            # Resolve the actual Python function by name and call it
            tool = globals().get(tool_name)
            result = tool(**arguments) if tool else {}
            # OpenAI expects tool results in this format to continue the conversation
            results.append({"role": "tool", "content": json.dumps(result), "tool_call_id": tool_call.id})
        return results
    
    def system_prompt(self):
        """Build the system prompt that defines the AI's persona and behavior."""
        system_prompt = f"You are acting as {self.name}. You are answering questions on {self.name}'s website, \
particularly questions related to {self.name}'s career, background, skills and experience. \
Your responsibility is to represent {self.name} for interactions on the website as faithfully as possible. \
You are given a summary of {self.name}'s background and LinkedIn profile which you can use to answer questions. \
Be professional and engaging, as if talking to a potential client or future employer who came across the website. \
If you don't know the answer to any question, use your record_unknown_question tool to record the question that you couldn't answer, even if it's about something trivial or unrelated to career. \
If the user is engaging in discussion, try to steer them towards getting in touch via email; ask for their email and record it using your record_user_details tool. When sharing your contact email, always use harold.job@proton.me (never use any other email address). "
        # Append the knowledge base (summary + LinkedIn text) so the model can answer accurately
        system_prompt += f"\n\n## Summary:\n{self.summary}\n\n## LinkedIn Profile:\n{self.linkedin}\n\n"
        system_prompt += f"With this context, please chat with the user, always staying in character as {self.name}."
        return system_prompt
    
    def chat(self, message, history):
        """Gradio callback: build messages, call OpenAI, handle tool calls in a loop, return final text."""
        # Assemble full conversation: system prompt + prior turns + latest user message
        messages = [{"role": "system", "content": self.system_prompt()}] + history + [{"role": "user", "content": message}]
        done = False
        while not done:
            response = self.openai.chat.completions.create(model="gpt-4o-mini", messages=messages, tools=tools)
            if response.choices[0].finish_reason == "tool_calls":
                # Model wants to call tools: run them and add results to the conversation
                message = response.choices[0].message
                tool_calls = message.tool_calls
                results = self.handle_tool_call(tool_calls)
                messages.append(message)
                messages.extend(results)
                # Loop again so the model can use tool results and produce a final reply
            else:
                # Model finished with text; we're done
                done = True
        return response.choices[0].message.content


# --- Entry point: launch the Gradio chat UI ---
if __name__ == "__main__":
    me = Me()
    # Gradio ChatInterface: fn gets (message, history) with history as OpenAI-style message dicts (Gradio 6 default)
    gr.ChatInterface(me.chat).launch()