hyperstack-py 1.5.2

Memory Hub for AI agents. Typed graph memory with episodic/semantic/working surfaces, decision replay, utility-weighted edges, git-style branching, and zero-friction markdown ingest. Includes can(), plan(), and auto_remember() for agent coordination. Zero LLM cost.

🃏 HyperStack Python SDK

Cloud memory for AI agents. Zero dependencies. 3 lines to integrate.

Install

pip install hyperstack-py

Quick Start

from hyperstack import HyperStack

hs = HyperStack("hs_your_key")

# Store a memory
hs.store("project-api", "API", "FastAPI 3.12 on AWS", stack="projects", keywords=["fastapi", "python"])

# Search memories
results = hs.search("python")

# List all cards
cards = hs.list()

# Delete a card
hs.delete("project-api")

# Get usage stats
stats = hs.stats()
print(f"Saving {stats['savings_pct']}% on tokens!")

# Auto-extract from conversation text
hs.ingest("Alice is a senior engineer. We decided to use FastAPI over Django.")

Why HyperStack?

• Zero dependencies — just Python stdlib
• No LLM costs — memory ops are free
• 94% token savings — ~350 tokens vs ~6,000 per message
• 30-second setup — get key at cascadeai.dev

API Reference

Method	Description
store(slug, title, body, stack, keywords)	Create/update a card
search(query)	Search cards
list(stack=None)	List all cards
get(slug)	Get one card
delete(slug)	Delete a card
stats()	Usage summary
ingest(text)	Auto-extract memories

Get a free key

→ cascadeai.dev — 50 cards free, no credit card.

License

MIT © CascadeAI