Master optimization system - APPLIES TO EVERY RESPONSE. Before responding, classify task complexity (simple question vs analysis vs coding). Use Haiku for simple/navigation/extraction/status. Use Sonnet ONLY for writing/analysis/planning/debugging. Monitor context size - if >50k tokens, recommend /compact. For automations, use scheduler wrapper. Never load full conversation history for simple tasks. Heartbeats always Haiku, single-line only. Never use Opus. This skill MUST run before every response to prevent 100k+ token bloat.
- Migrated all major components from Python scripts to modular JavaScript files for better integration and performance. - Introduced automatic context compaction with token tracking and summarization at 50,000 tokens. - Enhanced task routing logic with intelligent model selection and real-time cost prediction. - Improved scheduler with robust preflight/postflight hooks and fault-tolerant execution. - Added real-time dashboard for budget, task, and model usage monitoring. - Removed legacy scripts and static state files, streamlining the architecture.