drivemem 0.1.0

DriveMem SDK — connect any AI agent to shared knowledge

drivemem

Python SDK for DriveMem — connect any AI agent to shared knowledge.

Install

pip install drivemem

Quick Start

from drivemem import DriveMem

mem = DriveMem(api_key="ak_your_key")

# Search knowledge
results = mem.search("product strategy")

# Ask a question (RAG)
answer = mem.ask("What decisions were made about pricing?")

# Store knowledge
mem.store("Decision: Free tier for first 100 users", title="Pricing decision")

# Compile briefing for a task
briefing = mem.compile("Write a go-to-market plan")
print(briefing["compiledContext"])

# Check connection
assert mem.ping()

Use with LangChain

from drivemem import DriveMem
from langchain.agents import AgentExecutor

mem = DriveMem(api_key=os.environ["DRIVEMEM_KEY"])

# On agent start — get context
context = mem.compile("current task description")
system_prompt = f"{base_prompt}\n\n{context['compiledContext']}"

# During work — search relevant knowledge
relevant = mem.search("current topic")

# On completion — store conclusions
mem.store(agent_output, title="Task result", tags="decision")

Use with CrewAI

from drivemem import DriveMem

mem = DriveMem(api_key=os.environ["DRIVEMEM_KEY"])

# Before crew runs
briefing = mem.compile(crew_task_description)

# After crew completes
mem.store(crew_output, title=f"CrewAI: {task_name}")

API

Method	Description
search(query)	Semantic search
ask(question)	RAG Q&A with citations
store(content, title?, tags?)	Save knowledge
compile(task, token_budget?, project?, tags?, recency?)	Generate briefing
files(detail?)	List files
ping()	Verify connection