All Frameworks

Python SDK Quickstart

Add persistent memory to any Python AI agent in under 5 minutes.

1Install & Configure

Install the SDK from PyPI and set your API key.

pip install agenticmemory-sdk

export AGENTICMEMORY_API_KEY="amk_your_key_here"

2Store, Recall & Search

Create a space, store messages, search by relevance, and use key-value context.

from agenticmemoryai import AgenticMemory

# Initialize (reads AGENTICMEMORY_API_KEY from env)
memory = AgenticMemory()

# Create a memory space
space = memory.create_space("My Agent", "my-agent")
space_id = space["space"]["_id"]

# Store messages (short-term memory)
memory.store(space_id, role="user", content="The project deadline is June 15.")
memory.store(space_id, role="assistant", content="Got it, I'll track that deadline.")

# Recall recent messages
msgs = memory.recall(space_id, limit=20)
for m in msgs["messages"]:
    print(f"{m['role']}: {m['content']}")

# Search memory (Pro plan)
results = memory.search(space_id, "deadline", limit=10, scope="all")
print("Search hits:", len(results["results"]))

# Key-value context
memory.set_context(space_id, "user_prefs", {"theme": "dark", "lang": "en"}, ttl=3600)
prefs = memory.get_context(space_id, "user_prefs")
print("Prefs:", prefs)

# Long-term entries
memory.add_entry(space_id, type="summary", title="Sprint 4",
                 content="Delivered auth module on time", tags=["project"])
entries = memory.get_entries(space_id, limit=10, tags=["project"])

# Entities (people, concepts)
memory.add_entity(space_id, "Alice", role="engineer", knows=["Python", "Redis"])
entities = memory.get_entities(space_id, knows_contains="Python")

# Bootstrap: load everything at once for agent context
ctx = memory.bootstrap(space_id, entries=5, messages=20, semantic=True, query="deadline")
print("Context keys:", list(ctx.keys()))

3Expected Output

user: The project deadline is June 15.
assistant: Got it, I'll track that deadline.
Search hits: 2
Prefs: {'theme': 'dark', 'lang': 'en'}
Context keys: ['messages', 'entries', 'context', 'entities', 'shared', 'search']

---

Free tier: 1 space, 1K messages/month. Upgrade to Pro ($24.99/mo) for semantic search, entries, and entities.