skills.rest
kamahei
kamaheiCOMMUNITY
Explore all skills by this author →
GitHub

multi-agent-consensus

Coordinate 2-agent or 3-agent reasoning sessions with fixed participants and stop conditions.

Updated Mar 12, 2026
Search Tags:#workflow#multi-agent#consensus#coordination#scenarios#decision-support#safeguards
One-click install
npx skills add https://github.com/kamahei/agent-skills-lab --skill multi-agent-consensus
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: multi-agent-consensus
Source: https://github.com/kamahei/agent-skills-lab/tree/main/.github/skills/multi-agent-consensus
Command: npx skills add https://github.com/kamahei/agent-skills-lab --skill multi-agent-consensus

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Run the /multi-agent-consensus workflow for on-demand 2-agent or 3-agent reasoning and review with bounded refinement and explicit stop conditions. Use only when the user explicitly invokes /multi-agent-consensus. Do not use for generic mentions of multi-agent work, consensus, or second-opinion review without that exact skill invocation.

Core Features & Use Cases

  • On-demand 2-agent or 3-agent reasoning sessions with fixed participant sets.
  • Deterministic workflow with bounded refinement and explicit stop conditions.
  • Clear governance of model selections and slot counting to ensure independent participants.

Quick Start

Invoke the /multi-agent-consensus command to start a coordinated 2-agent or 3-agent run with predetermined participants.

Frequently Asked Questions about multi-agent-consensus

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I coordinate multi-agent reasoning for structured debate?▼

Multi-agent reasoning coordination is handled by invoking the /multi-agent-consensus command, which triggers a deterministic workflow with fixed participant sets and explicit stop conditions for structured debate.

Can I run a 2-agent or 3-agent decision support session with bounded refinement?▼

Yes, you can run on-demand 2-agent or 3-agent decision support sessions. The workflow enforces default agent counts when unspecified and applies strict participant selection rules to ensure bounded refinement.

What is the best way to set up scenario evaluation with multiple AI participants?▼

Scenario evaluation with multiple participants is best initiated by explicitly invoking the /multi-agent-consensus command. This starts a coordinated run with predetermined model selections and clear stopping criteria.

Does the multi-agent consensus workflow require explicit invocation to start?▼

Yes, the multi-agent consensus workflow requires explicit invocation via the /multi-agent-consensus command. It should not be triggered by generic mentions of consensus, multi-agent work, or second-opinion review.

How are participant slots counted during a multi-agent reasoning session?▼

Participant slot counting governs model selections to ensure independent participants. The workflow enforces strict participant selection rules and applies default agent counts when not explicitly specified by the user.

When should I not use the multi-agent consensus workflow?▼

You should not use the multi-agent consensus workflow for generic mentions of multi-agent work, consensus, or second-opinion review. It is strictly reserved for runs explicitly triggered by the /multi-agent-consensus command.

Related Skills

Explore More Software Engineering Skills→
rocklambrosrocklambros
2

scaffolding-llm-eval-harness

Scaffold a reproducible LLM evaluation harness with versioned prompts and JSONL results.

Community
Advanced
princespaghettiprincespaghetti

council-consultation

Generate 4-5 expert perspectives with analysis and synthesis for complex decisions.

Community
Advanced
ollieb89ollieb89

langchain-architecture

Design LangChain AI applications with agents, memory, and tool integration.

Community
Advanced
kamaheikamahei

multi-agent-isolated-synthesis

Coordinate two or three agents in isolated workspaces to synthesize a consensus result.

Community
Advanced