What problem does it solve? Getting a well-rounded answer to a complex question often requires multiple perspectives and current web evidence, but a single model response can be one-sided or outdated. This Skill convenes a fixed three-agent General Council (generalist, skeptic, domain expert) that researches the web first, deliberates in parallel, reconciles disagreements, and delivers a synthesized advisory answer. ## Core Features & Use Cases - Curated Web Research: Runs 1-3 targeted web_search queries upfront and compiles a RESEARCH CONTEXT block shared with all council members. - Parallel Multi-Agent Deliberation: Dispatches generalist, skeptic, and domain expert agents in parallel, then routes disagreements back for one targeted reconciliation round. - Honest Synthesis: Leads with consensus, explicitly acknowledges unresolved disagreements, and cites the strongest sources without inventing claims. - Use Case: A user runs /swarm council Should we migrate our monolith to microservices? and receives a concise, source-cited answer reflecting three independent expert viewpoints plus flagged disagreements. ## Quick Start Ask the agent to run /swarm council followed by your question to get a synthesized multi-expert advisory answer.