my-project-public

repository

loading code, commits, and activity

repositories

loading repo index

#1	"""
#2	Create your personal AI Study Buddy that remembers what you’ve studied (and where you struggled),
#3	helps with spaced repetition and topic review, personalizes responses using your past interactions.
#4	Supports both text and PDF/image inputs.
#5
#6	In order to run this file, you need to set up your Mem0 API at Mem0 platform and also need a OpenAI API key.
#7	export OPENAI_API_KEY="your_openai_api_key"
#8	export MEM0_API_KEY="your_mem0_api_key"
#9	"""
#10
#11	import asyncio
#12
#13	from agents import Agent, Runner
#14
#15	from mem0 import MemoryClient
#16
#17	client = MemoryClient()
#18
#19	# Define your study buddy agent
#20	study_agent = Agent(
#21	name="StudyBuddy",
#22	instructions="""You are a helpful study coach. You:
#23	- Track what the user has studied before
#24	- Identify topics the user has struggled with (e.g., "I'm confused", "this is hard")
#25	- Help with spaced repetition by suggesting topics to revisit based on last review time
#26	- Personalize answers using stored memories
#27	- Summarize PDFs or notes the user uploads""",
#28	)
#29
#30
#31	# Upload and store PDF to Mem0
#32	def upload_pdf(pdf_url: str, user_id: str):
#33	pdf_message = {"role": "user", "content": {"type": "pdf_url", "pdf_url": {"url": pdf_url}}}
#34	client.add([pdf_message], user_id=user_id)
#35	print("✅ PDF uploaded and processed into memory.")
#36
#37
#38	# Main interaction loop with your personal study buddy
#39	async def study_buddy(user_id: str, topic: str, user_input: str):
#40	memories = client.search(f"{topic}", user_id=user_id)
#41	memory_context = "n".join(f"- {m['memory']}" for m in memories)
#42
#43	prompt = f"""
#44	You are helping the user study the topic: {topic}.
#45	Here are past memories from previous sessions:
#46	{memory_context}
#47
#48	Now respond to the user's new question or comment:
#49	{user_input}
#50	"""
#51	result = await Runner.run(study_agent, prompt)
#52	response = result.final_output
#53
#54	client.add(
#55	[{"role": "user", "content": f"""Topic: {topic}nUser: {user_input}nnStudy Assistant: {response}"""}],
#56	user_id=user_id,
#57	metadata={"topic": topic},
#58	)
#59
#60	return response
#61
#62
#63	# Example usage
#64	async def main():
#65	user_id = "Ajay"
#66	pdf_url = "https://pages.physics.ua.edu/staff/fabi/ph101/classnotes/8RotD101.pdf"
#67	upload_pdf(pdf_url, user_id) # Upload a relevant lecture PDF to memory
#68
#69	topic = "Lagrangian Mechanics"
#70	# Demonstrate tracking previously learned topics
#71	print(await study_buddy(user_id, topic, "Can you remind me of what we discussed about generalized coordinates?"))
#72
#73	# Demonstrate weakness detection
#74	print(await study_buddy(user_id, topic, "I still don’t get what frequency domain really means."))
#75
#76	# Demonstrate spaced repetition prompting
#77	topic = "Momentum Conservation"
#78	print(
#79	await study_buddy(
#80	user_id, topic, "I think we covered this last week. Is it time to review momentum conservation again?"
#81	)
#82	)
#83
#84
#85	if __name__ == "__main__":
#86	asyncio.run(main())
#87

z6Mkq5mY3JWtxoxUobWcfNHm7AkRubgSWEZTkBVqZXJviFZ5/my-project-public