Persistent graph memory for OpenClaw. Two-minute install. MCP-native.
OpenClaw's conversational memory works within a session. For agents that need to remember yesterday's research, last week's customer context, or cross-reference information across threads, a persistent memory layer is the missing piece.
Works out of the box with any OpenClaw configuration that supports MCP servers. No wrappers, no glue code.
Your OpenClaw agent can traverse real business relationships, not just retrieve chunks.
No vectors, no hallucinated recall, no embedding costs. Query and get exact results.
Every memory carries a timestamp, a source, and an audit trail. Full audit trail with authority weights.
Bring OpenClaw, bring any model, bring your data. Graphory is the memory layer, not the reasoning layer.
Install the SDK, log in, and register the MCP server with OpenClaw.
Step 1 - Install and authenticate
pip install graphory graphory login
Step 2 - Register Graphory in your OpenClaw MCP servers config
{
"mcpServers": {
"graphory": {
"url": "https://api.graphory.io/mcp",
"headers": {
"Authorization": "Bearer gs_ak_your_key"
}
}
}
}Illustrative snippet. Use the exact MCP config format from your OpenClaw release - see the OpenClaw docs for the current schema.
Real queries your OpenClaw agent can run once Graphory is connected.
"Remember that Sarah from Acme mentioned pricing was their blocker in March."
"Show me everything my agent learned about Project X this quarter."
"Who have I interacted with at companies in the fintech vertical?"
Deterministic extraction and retrieval, benchmarked against public datasets.
0.9107 accuracy
LongMemEval (agent + MCP)
0.8667 accuracy
LoCoMo-MC10 (agent + MCP)
0.900 F1
BizLineItemBench
100K nodes, unlimited connections, full MCP and REST API access. Scale up only when your agent accumulates enough memory to need it.
Free up to 100K nodes. No credit card.