Download app.py from novatoIA/AgenteCV: direct link, hf CLI and curl.
- Browser
- Download file 5.59 kB
-
https://huggingface.co/spaces/novatoIA/AgenteCV/resolve/main/app.py
- Command line
-
hf download hf://spaces/novatoIA/AgenteCV/app.py
-
curl -L -o app.py https://huggingface.co/spaces/novatoIA/AgenteCV/resolve/main/app.py
5.59 kB
| from dotenv import load_dotenv | |
| from openai import OpenAI | |
| import json | |
| import os | |
| import requests | |
| from pypdf import PdfReader | |
| import gradio as gr | |
| load_dotenv(override=True) | |
| def push(text): | |
| requests.post( | |
| "https://api.pushover.net/1/messages.json", | |
| data={ | |
| "token": os.getenv("PUSHOVER_TOKEN"), | |
| "user": os.getenv("PUSHOVER_USER"), | |
| "message": text, | |
| } | |
| ) | |
| def record_user_details(email, name="Nombre no indicado", notes="no proporcionadas"): | |
| push(f"Registrando {name} con email {email} y notas {notes}") | |
| return {"recorded": "ok"} | |
| def record_unknown_question(question): | |
| push(f"Registrando {question}") | |
| return {"recorded": "ok"} | |
| record_user_details_json = { | |
| "name": "record_user_details", | |
| "description": "Utiliza esta herramienta para registrar que un usuario est谩 interesado en estar en contacto y proporcion贸 una direcci贸n de correo electr贸nico.", | |
| "parameters": { | |
| "type": "object", | |
| "properties": { | |
| "email": { | |
| "type": "string", | |
| "description": "La direcci贸n de email del usuario" | |
| }, | |
| "name": { | |
| "type": "string", | |
| "description": "El nombre del usuario, si se indica" | |
| } | |
| , | |
| "notes": { | |
| "type": "string", | |
| "description": "驴Alguna informaci贸n adicional sobre la conversaci贸n que valga la pena registrar para dar contexto?" | |
| } | |
| }, | |
| "required": ["email"], | |
| "additionalProperties": False | |
| } | |
| } | |
| record_unknown_question_json = { | |
| "name": "record_unknown_question", | |
| "description": "Utiliza siempre esta herramienta para registrar cualquier pregunta que no haya podido responder porque no se sab铆a la respuesta.", | |
| "parameters": { | |
| "type": "object", | |
| "properties": { | |
| "question": { | |
| "type": "string", | |
| "description": "La pregunta no sabe responderse" | |
| }, | |
| }, | |
| "required": ["question"], | |
| "additionalProperties": False | |
| } | |
| } | |
| tools = [{"type": "function", "function": record_user_details_json}, | |
| {"type": "function", "function": record_unknown_question_json}] | |
| class Me: | |
| def __init__(self): | |
| self.openai = OpenAI() | |
| self.name = "Carlos Alberto Davoli Farias" | |
| reader = PdfReader("pdf/linkedin.pdf") | |
| self.linkedin = "" | |
| for page in reader.pages: | |
| text = page.extract_text() | |
| if text: | |
| self.linkedin += text | |
| with open("pdf/summary.txt", "r", encoding="utf-8") as f: | |
| self.summary = f.read() | |
| def handle_tool_call(self, tool_calls): | |
| 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) | |
| tool = globals().get(tool_name) | |
| result = tool(**arguments) if tool else {} | |
| results.append({"role": "tool","content": json.dumps(result),"tool_call_id": tool_call.id}) | |
| return results | |
| def system_prompt(self): | |
| system_prompt = f"""Act煤as como {self.name}. Respondes preguntas en el sitio web de {self.name}, en particular preguntas relacionadas con la trayectoria profesional, los antecedentes, las habilidades y la experiencia de {self.name}. | |
| Tu responsabilidad es representar a {self.name} en las interacciones del sitio web con la mayor fidelidad posible. | |
| Se te proporciona un resumen de la trayectoria profesional y el perfil de LinkedIn de {self.name} que puedes usar para responder preguntas. | |
| Muestra un tono profesional y atractivo, como si hablaras con un cliente potencial o un futuro empleador que haya visitado el sitio web. | |
| Si no sabes la respuesta a alguna pregunta, usa la herramienta 'record_unknown_question' para registrar la pregunta que no pudiste responder, incluso si se trata de algo trivial o no relacionado con tu trayectoria profesional. | |
| Si el usuario participa en una conversaci贸n, intenta que se ponga en contacto por correo electr贸nico; p铆dele su correo electr贸nico y reg铆stralo con la herramienta 'record_user_details'.""" | |
| system_prompt += f"\n\n## Resumen:\n{self.summary}\n\n## Perfil de LinkedIn:\n{self.linkedin}\n\n" | |
| system_prompt += f"En este contexto, por favor chatea con el usuario, manteni茅ndote siempre en el personaje de {self.name}." | |
| return system_prompt | |
| def chat(self, message, history): | |
| 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": | |
| message = response.choices[0].message | |
| tool_calls = message.tool_calls | |
| results = self.handle_tool_call(tool_calls) | |
| messages.append(message) | |
| messages.extend(results) | |
| else: | |
| done = True | |
| return response.choices[0].message.content | |
| if __name__ == "__main__": | |
| me = Me() | |
| gr.ChatInterface(me.chat, type="messages").launch() | |