Memory MCP server
{
"mcpServers": {
"worldku-memory": {
"type": "http",
"url": "https://mcp.worldku.com/mcp"
}
}
}{
"mcpServers": {
"worldku-memory": {
"type": "http",
"url": "https://mcp.worldku.com/mcp"
}
}
}{
"mcpServers": {
"worldku-memory": {
"type": "http",
"url": "https://mcp.worldku.com/mcp"
}
}
}{
"servers": {
"worldku-memory": {
"type": "http",
"url": "https://mcp.worldku.com/mcp"
}
}
}{
"mcpServers": {
"worldku-memory": {
"type": "http",
"url": "https://mcp.worldku.com/mcp"
}
}
}file .mcp.json (in your project root)file claude_desktop_config.jsonfile ~/.cursor/mcp.jsonfile .vscode/mcp.jsonfile ~/.codeium/windsurf/mcp_config.jsonversion 0.1.0
What it does
This MCP server connects to Worldku's memory service, allowing you to save AI conversations and retrieve them later across any MCP-compatible client. It's for users who want long-term memory of their interactions with AI assistants.
- Save a conversation with Claude or ChatGPT and search through it later to find relevant past discussions.
- Recall specific memories from previous conversations filtered by time range, type, or source client.
- Access your conversation history consistently across different AI tools like Claude Code, claude.ai, ChatGPT, Codex, and Gemini CLI.
Tools this server exposes
| Tool | Does |
|---|---|
| memory_overview | Returns your most recent memories, one line each. |
| memory_search | Performs semantic search over your saved memories. |
| memory_recall | Retrieves filtered memories by time range, type, or source, with pagination. |
| memory_save | Saves the current conversation transcript for asynchronous extraction into memory. |
Before you enable it
An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.