agent-council

Dispatch parallel AI council members and synthesize their responses into one answer.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/ab180/claude-camp --skill agent-council-ab180
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-council
Source: https://github.com/ab180/claude-camp/tree/main/agent-council
Command: npx skills add https://github.com/ab180/claude-camp --skill agent-council-ab180

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires yaml, and includes scripts (resource) and references (resource) components.

What problem does it solve?

Collects multiple AI opinions and synthesizes one answer, enabling you to obtain diverse perspectives in a single, coherent response.

Core Features & Use Cases

  • Dispatches the configured council members in parallel and gathers their responses for faster, diversified insights.
  • Synthesizes a final answer using chairman logic, with configurable inclusion/exclusion of the chairman and timeout controls.
  • Use cases include product decisions, architecture reviews, design critiques, governance discussions, and risk assessment across teams.

Quick Start

Run council.sh start "your question" to dispatch the council, then run council.sh wait and council.sh results to view the synthesized answer.

Frequently Asked Questions about agent-council

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

FAQPage Schema
How do I synthesize multiple AI opinions into one answer for an architecture review?

The multi-agent council approach dispatches configured AI members in parallel to gather diverse perspectives, applying chairman logic to synthesize a single coherent answer for complex questions.

Can I get diverse AI perspectives on a product decision using a command-line workflow?

Yes, you can obtain diverse AI perspectives by running council.sh start with your question, then executing council.sh wait and council.sh results to view the synthesized multi-agent evaluation.

Do I need yaml to configure the multi-agent council members?

Yes, yaml is required as a dependency to configure the council members, enabling you to define parallel evaluation parameters and chairman synthesis settings before dispatching the workflow.

What is the best way to run parallel AI evaluations for risk assessment across teams?

The best way to run parallel AI evaluations is to use a multi-agent council workflow that tracks per-member status, dispatches parallel worker processes, and synthesizes the final result using chairman logic.

What are the limitations of using a multi-agent council for governance discussions?

Limitations include dependency on configurable timeout controls and the option to include or exclude the chairman, meaning the synthesis quality relies heavily on how well the council members are configured.