Spawn expert agents to debate or poll and synthesize trade-offs.
npx skills add https://github.com/hungv47/meta-skills --skill debate-agents
Please help me install this Agent Skill. Skill: debate-agents Source: https://github.com/hungv47/meta-skills/tree/main/skills/meta/debate-agents Command: npx skills add https://github.com/hungv47/meta-skills --skill debate-agents
Enables framing and testing of hard decisions by running multiple expert perspectives through debate or poll, producing a structured synthesis that surfaces trade-offs and disagreements.
Frame the decision clearly, choose the mode (debate or poll), and invoke debate-agents to obtain a reproducible synthesis.
High-intent search queries and answers about installing and using this skill.
Multi-agent debate improves decision support by spawning 3-5 expert perspectives to argue high-stakes questions, surfacing trade-offs and disagreements. This structured risk analysis produces a synthesis with recommended actions and unresolved risks for robust evaluation.
Debate mode spawns 3-5 distinct agents to argue opposing viewpoints, while poll mode gathers input from 10+ agents. Both generate structured consensus outputs, but debate emphasizes surfacing deep trade-offs whereas poll captures broader multi-perspective input.
Frame your decision clearly, choose debate or poll mode, and invoke the system to obtain a reproducible synthesis. The structured output includes participants, consensus, key disagreements, recommended action, and unresolved risks for audit-friendly records.
Yes, multi-agent debate supports sub-routine invocation by other skills, returning inline synthesis directly to the caller's flow. This enables integration into larger decision pipelines like discover, prioritize, or architect-system as a sub-routine step.
No external dependencies are required to run multi-agent debate. The standalone mode writes dated artifact records autonomously, and sub-routine mode returns inline synthesis without needing external libraries or frameworks.
Use standalone mode when you need dated artifact records for independent audit and review. Use sub-routine mode when integrating consensus analysis directly into a larger decision pipeline to return inline synthesis back to the calling workflow.
magnus919Orchestrate multi-agent structured debates to analyze complex decisions and surface hidden assumptions.
vasilievyakovCoordinates five expert directors to evaluate a project and synthesize consensus and disagreements.
rysweetFacilitate structured multi-perspective debates for architectural decisions.
syxscottConvert text to speech using the ElevenLabs API with a macOS say-style interface.
