SuperMemory runs on your own machine — the client starts it, so there is no endpoint to ping. 58 installs a week from pypi. Last commit 14 Jun 2026.
MCP-first agent learning: capture failures, validate lessons, retrieve context, improve.
Today is the operative word: we check SuperMemory every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add supermemory -- uvx supermemory-agent
{
"mcpServers": {
"supermemory": {
"args": [
"supermemory-agent"
],
"command": "uvx"
}
}
}
[mcp_servers.supermemory]
command = "uvx"
args = ["supermemory-agent"]
{
"mcpServers": {
"supermemory": {
"args": [
"supermemory-agent"
],
"command": "uvx"
}
}
}
{
"mcpServers": {
"supermemory": {
"args": [
"supermemory-agent"
],
"command": "uvx"
}
}
}
This one needs environment variables set before it will start:
SUPERMEMORY_STORAGE_PATH (Local directory for .supermemory file storage), UALL_DATA_DIR (Legacy alias for SUPERMEMORY_STORAGE_PATH).
The author declared them in the registry entry; get the values from the project itself.
Failure-memory layer for coding agents: search evidence-rated failure lessons by error text.
Persistent memory for AI agents. Store context, retrieve it semantically.
Persistent memory for AI coding agents. Hybrid search, auto-session capture, context routing.
Local-first, formally verified memory for AI agents: consult/learn/related/list over MCP.
Memory that reasons: continual learning for stateful agents. Better context, fewer tokens.
Context-aware code retrieval MCP server — ranked, gated results for AI coding agents
Local-first memory brain for AI coding agents: hooks capture sessions, beliefs injected per prompt.
Persistent memory and cross-session learning for AI coding assistants (hosted remote MCP).
Answers built from our own checks of this server.