multi-agent-orchestrationlisted

# Multi-Agent Orchestration Skill Route and delegate tasks to the most appropriate AI provider based on task characteristics and provider capabilities. ## Variables | Variable | Default | Description | |----------|---------|-------------| | ENABLED_CLAUDE | true | Enable Claude Code as provider | | ENABLED_OPENAI | true | Enable OpenAI/Codex as provider | | ENABLED_GEMINI | true | Enable Gemini as provider | | ENABLED_CURSOR | true | Enable Cursor as provider | | ENABLED_OPENCODE | true | Enable OpenCode as provider | | ENABLED_OLLAMA | true | Enable local Ollama as provider | | DEFAULT_PROVIDER | claude | Fallback when routing is uncertain | | CHECK_COST_STATUS | true | Check usage before delegating | ## Instructions **MANDATORY** - Follow the Workflow steps below in order. Do not skip steps. - Before delegating, understand the task characteristics - Use the model-discovery skill for current model names - Check cost/usage status before high-volume delegation ## Quick Decision Tree ``` What type of task is this? │ ├─ Needs conversation history? ─────────► Keep in Claude (no delegation) │ ├─ Needs sandboxed execution? ──────────► OpenAI/Codex │ ├─ Large context (>100k tokens)? ───────► Gemini │ ├─ Multimodal (images/video)? ──────────► Gemini │ ├─ Needs web search? ───────────────────► Gemini │ ├─ Quick IDE edit? ─────────────────────► Cursor │ ├─ Privacy required / offline? ─────────► Ollama │ ├─ Provider-agnostic fallback? ─────────► OpenCode │ └─ General reasoning /