🧠 memX
Real-time shared memory for multi-agent LLM systems: no chat, no controller.
No setup. Free to use. Just shared memory between agents.
⚡ Why memX?
- Real-time shared key-value store
- Pub/Sub updates across agents
- JSON Schema enforcement
- API-key-based access control
🚀 Quickstart
- Install SDK:
pip install memx_sdk - Get your API key
- Start coding:
from memx_sdk import memxContext
ctx = memxContext(api_key="your_api_key")
ctx.set_schema("agent:goal", {
"type": "object",
"properties": {"x": {"type": "number"}, "y": {"type": "number"}},
"required": ["x", "y"]
})
ctx.set("agent:goal", {"x": 1, "y": 7})
print(ctx.get("agent:goal"))
🧬 Example: Multi Agents, One Memory
🌐 Use Cases
- Autonomous research agents
- LangGraph / CrewAI memory plugins
- LLM workflows with persistent state