Memstate + Grok SDK

pip install xai-sdk

export XAI_API_KEY="xai-..."
export MEMSTATE_API_KEY="mst_..."

"""
Memstate AI + Grok (xAI SDK)
Requires: pip install xai-sdk
Env: XAI_API_KEY, MEMSTATE_API_KEY

Grok uses REMOTE MCP over HTTP/SSE. The Memstate hosted MCP endpoint
is available at https://mcp.memstate.ai/sse — no local server needed.
"""
import os
from xai_sdk import Client
from xai_sdk.chat import user
from xai_sdk.tools import mcp

client = Client(api_key=os.environ["XAI_API_KEY"])

# Grok connects to Memstate via the hosted remote MCP endpoint
chat = client.chat.create(
    model="grok-3-latest",
    tools=[
        mcp(
            server_url="https://mcp.memstate.ai/sse",
            server_label="memstate",
            server_description="Persistent versioned memory for AI agents",
            authorization=f"Bearer {os.environ['MEMSTATE_API_KEY']}",
        ),
    ],
)

chat.append(
    user(
        'Save a memory: memstate_remember(project_id="demo", '
        'content="## Grok Integration\\nMemstate connected via xAI remote MCP.", '
        'source="agent"). Then call memstate_get(project_id="demo").'
    )
)

for response, chunk in chat.stream():
    if chunk.content:
        print(chunk.content, end="", flush=True)

print()

I'm onboarding Memstate AI memory for this project. Please:
1. Analyze this codebase and write a concise high-level architecture overview in markdown.
2. Save it with: memstate_remember(project_id="<your_project>", content="<the markdown>", source="agent")
3. Then call memstate_get(project_id="<your_project>") and show me the memory tree.