What problem does it solve? AI agents lose all context when a session restarts, forcing users to repeat themselves and re-explain projects, people, and preferences. This Skill provides a file-based memory system that persists knowledge across sessions using Tiago Forte's PARA method. ## Core Features & Use Cases - Three-Layer Memory: A PARA-organized knowledge graph (projects, areas, resources, archives) with atomic YAML facts, daily timeline notes, and a tacit knowledge file capturing user patterns. - Memory Decay and Synthesis: Facts are tracked with access counts and recency tiers (hot, warm, cold), and weekly synthesis rewrites summaries so stale information drops out of context without being deleted. - Semantic Recall via qmd: Retrieve past context with vector search, BM25 keyword search, or hybrid reranking instead of grepping files. - Use Case: After a user mentions a client three times, create an entity folder under areas/companies, store durable facts in items.yaml, and recall them weeks later with a qmd semantic query. ## Quick Start Ask the agent to remember a fact about a current project and store it in the appropriate PARA entity folder.
