memweave is a zero-infrastructure, async-first Python library that gives AI agents persistent, searchable memory — stored as plain Markdown files

Temporal memory system for AI assistants with human-like forgetting curves. All data stored locally in human-readable formats: JSONL for short-term memory, Markdown (Obsidian-compatible) for long-term. Memories naturally decay unless reinforced. Features knowledge graphs, smart prompting, and MCP server integration for Claude.

The open-source benchmark for LLM memory decay. Measure how Naive, RAG, Chunked RAG, Cascading, and SummaryMemory degrade over 100 conversation turns. Ebbinghaus forgetting curves, 5-provider LLM eval, multi-seed CI. No API key needed.

Persistent associative memory for LLM agents that learns what connects to what through repeated exposure

Turn your Obsidian vault into a persistent, plain-text memory system that improves AI assistants’ understanding across sessions.

Adaptive memory retention for RAG systems. VDD dynamically adjusts decay based on semantic drift. 42 reproducible experiments, 120 real-world facts, 3 domains. Never the worst method under unknown drift.

A lightweight context engine for AI agents. Ingest events, build semantic context, query with natural language. Zero config.

Native temporal decay scoring for vector database retrieval — no post-retrieval hacks.

Provide a structured Obsidian vault template that integrates Claude for managing notes and syncing projects using a clear folder system.

Project Yui — portable identity preservation for AI minds. Crystal spec, patterns, and reference implementation. Apache 2.0.