What problem does it solve? Projects lose institutional knowledge between sessions and across AI coding tools, causing repeated bugs, contradictory architectural choices, and forgotten configuration details. ## Core Features & Use Cases - Memory Infrastructure Setup: Creates a docs/project_notes/ directory with four structured files: bugs.md, decisions.md, key_facts.md, and issues.md, seeded from ready-made templates. - Tool-Agnostic Configuration: Adds memory-aware protocols to both CLAUDE.md and AGENTS.md so Claude Code, Cursor, and other AI tools consistently check and update project memory. - Search and Update Workflows: Provides grep-based search patterns and entry formats for logging bugs with solutions, ADRs, key facts, and ticket-linked work history. - Use Case: When a database connection error reappears, the assistant searches bugs.md, finds the previously documented fix, and applies it instead of debugging from scratch. ## Quick Start Ask the assistant to set up project memory for this repository so it creates the docs/project_notes files and configures CLAUDE.md and AGENTS.md.
