What problem does it solve? AI agents lose all context between sessions, forcing users to repeat facts, preferences, and project history. This Skill provides a file-based memory system that persists durable knowledge across sessions using Tiago Forte's PARA method. ## Core Features & Use Cases - Three-Layer Memory: A PARA knowledge graph (projects, areas, resources, archives) for entities, daily notes for raw timelines, and a tacit knowledge file for user patterns and preferences. - Atomic Fact Schema with Decay: Facts stored in items.yaml carry status, timestamps, and access counts; weekly synthesis rewrites summaries using hot/warm/cold recency tiers without ever deleting data. - Semantic Recall via qmd: Search memory with vector similarity, BM25 keyword search, and reranking instead of grepping files. - Use Case: A user mentions a coworker across several sessions. The agent creates an entity under areas/people, logs atomic facts, and later recalls the relationship instantly when planning a project involving that person. ## Quick Start Ask the agent to remember a fact about a person or project, and it will write it to the appropriate PARA entity file and daily note for future recall.
