Save 60-90% on LLM token costs with intelligent memory compression. agent-memory automatically compresses tool output into structured observations, shares context across agents via a memory bus, and injects only relevant memory into each prompt — keeping your token budget under control. Features: auto-compression hooks (PostToolUse), token-budgeted context injection (SessionStart), semantic search via FTS5, cross-session persistence, and a token economics dashboard showing compression ratios and savings.