Projectmem
Provides a local-first memory and judgment layer for AI coding agents, capturing development history to reduce token usage and prevent repeated errors.
Projectmem offers a local-first memory and judgment layer designed for AI coding agents, transforming them from amnesiac to experienced by capturing crucial development context. It solves the problem of AI agents repeatedly forgetting past decisions, failed attempts, and debugging history, which leads to high token consumption. By integrating directly into your development workflow via Git hooks and a real-time file watcher, projectmem logs issues, decisions, and cross-project gotchas, then distills this memory to efficiently inject token-budgeted context into future AI sessions, thereby making AI interactions more efficient and effective.