Provides persistent memory for AI tools, enabling every session to leverage context from all previous interactions.
Sponsored
Memento empowers your AI tools with a crucial persistent memory, ensuring every session builds upon the context established in prior interactions. It operates entirely on your machine, connecting to any MCP-compatible AI application to construct a dynamic knowledge graph from your conversations. This graph captures entities, relationships, decisions, and crucial context, eliminating the need for repetitive explanations and enabling your AI to remember and learn over time. Without relying on cloud services, API keys, or subscriptions, Memento keeps your data private and local.
Características Principales
019 GitHub stars
02Intuitive web UI with dashboard, graph visualizer, and one-click integration setup
03Auto-extracts 22 entity types and maps 44 relationship types into a knowledge graph
04Manages memory lifecycle with states, decay scoring, and access-frequency boosting
05Runs entirely offline with local Ollama support
06Hybrid search combining full-text, semantic vector, and RRF ranking
Casos de Uso
01Establish shared team memory for engineering teams, making decisions and conventions queryable by anyone.
02Empower individual AI assistants (e.g., Claude Code, Cursor) with cross-session context recall and decision storage.
03Automate context management for AI tools, allowing them to proactively store and recall information without explicit prompting.
