Loading connector details…
Loading connector details…
Choose a unique username to continue using AgentHotspot
by meetdhanani17 • Data & Databases
A TypeScript-based MCP server that provides project-scoped, knowledge-graph memory (entities, relations, observations) with disk persistence for LLMs and agents.
Store and retrieve structured project-specific memory (entities, relations, observations) across sessions.
A simple, disk-persistent knowledge graph backend accessible via MCP tools and usable in local or Docker environments.
Cross-project knowledge sharing and migration of memories between projects or environments.
xgmem is a Model Context Protocol (MCP) server written in TypeScript that stores and serves structured memory—entities, relations, and observations—per project. It enables CRUD operations via MCP tools, supports cross-project knowledge sharing, and persists memory to disk (memory.json) with optional Docker deployment. The server is intended to be integrated into agent and LLM workflows to provide queryable, project-specific knowledge graphs and durable memory storage.
Save observations for a project to a separate file in the format { projectId, observations: [{ entityName, contents: [...] }] }.
Get all observations saved in the separate file for a given projectId.
Add new observations to existing entities in the project knowledge graph.
List all projects with stored memory
Create multiple new entities in the project knowledge graph
Scores are informational only and provided “as is” without warranty. AgentHotspot assumes no liability for actions taken based on these ratings.