Run a multi-LLM council task with adversarial debate
Run the LLM Council with the specified subagent and task. > **Requires:** `pip install the-llm-council` and at least one API key configured. ## Subagents - `implementer` - Feature code, bug fixes - `reviewer` - Code review, security audit - `architect` - System design, APIs - `researcher` - Technical research - `planner` - Roadmaps, execution plans - `assessor` - Build vs buy decisions - `red-team` - Security threat analysis - `test-designer` - Test suite design - `shipper` - Release notes - `router` - Task classification ## Execution