Use the hosted instance
The fastest path — no setup required:Copy
Ask AI
export PLYRA_SERVER_URL=https://plyra-memory-server.politedesert-a99b9eaf.centralindia.azurecontainerapps.io
export PLYRA_API_KEY=plm_live_your_key_here
Self-host with Docker
Copy
Ask AI
docker run -d \
--name plyra-memory-server \
-p 7700:7700 \
-e PLYRA_ADMIN_API_KEY=your_strong_admin_key \
-v plyra-data:/data \
ghcr.io/plyraai/plyra-memory-server:latest
http://localhost:7700.
Verify it’s running
Copy
Ask AI
curl http://localhost:7700/health
# → {"status": "ok", "version": "..."}
Create your first API key
Copy
Ask AI
curl -X POST http://localhost:7700/admin/keys \
-H "Authorization: Bearer your_strong_admin_key" \
-H "Content-Type: application/json" \
-d '{
"workspace_id": "my-workspace",
"label": "development",
"env": "test"
}'
# → {"key": "plm_test_...", "key_id": "...", "workspace_id": "..."}
The key is returned once. Copy it immediately — it cannot be retrieved again.
Connect plyra-memory
Copy
Ask AI
export PLYRA_SERVER_URL=http://localhost:7700
export PLYRA_API_KEY=plm_test_your_key_here
Copy
Ask AI
from plyra_memory import Memory
import asyncio
async def main():
async with Memory(agent_id="my-agent") as mem:
await mem.remember("server mode is working")
ctx = await mem.context_for("is it working")
print(ctx.content)
asyncio.run(main())
Docker Compose
Copy
Ask AI
version: "3.9"
services:
plyra-memory-server:
image: ghcr.io/plyraai/plyra-memory-server:latest
ports:
- "7700:7700"
environment:
PLYRA_ADMIN_API_KEY: ${PLYRA_ADMIN_API_KEY}
volumes:
- plyra-data:/data
volumes:
plyra-data:
Copy
Ask AI
PLYRA_ADMIN_API_KEY=your_strong_key docker compose up -d