minimax

Multi-agent orchestrator for Claude Code. Track work with convoys, sling to polecats. The Cognition Engine for AI-powered software factories.

Self-evolving research methodology: 18 universal principles for LLM-driven investigation, distilled from a 376-turn session with 1 positive + 17 null campaigns. Covers epistemic foundations (causality, nulls, record-keeping), investigation patterns (multi-lens agents, adversarial gates, per-trade enrichment), decision layer (constraint-pivot, supersede-not-rewrite), and emergent resurrection (failure archive with resurrect_if conditions).

This skill should be used when the model's ROLE_TYPE is orchestrator and needs to delegate tasks to specialist sub-agents. Provides scientific delegation framework ensuring world-building context (WHERE, WHAT, WHY) while preserving agent autonomy in implementation decisions (HOW). Use when planning task delegation, structuring sub-agent prompts, or coordinating multi-agent workflows.

HelloAGENTS — The orchestration kernel that makes any AI CLI smarter. Adds intelligent routing, unified QA gates, safety guards, and notifications.

Ultra-compressed communication mode. Cuts 65% of output tokens (measured) while keeping full technical accuracy by speaking like a caveman.

Frontend design skill for UI/UX implementation