Stats
Actions
Tags
Help us improve
Share bugs, ideas, or general feedback.
From llm-council-plugin
Displays detailed help for the LLM Council plugin: overview, commands, three-phase workflow, council members, requirements, examples, quorum rules, configuration, and troubleshooting.
npx claudepluginhub xrf9268-hue/llm-council-plugin --plugin llm-council-pluginHow this command is triggered — by the user, by Claude, or both
Slash command
/llm-council-plugin:council-helpclaude-haiku-4-5-20251001The summary Claude sees in its command listing — used to decide when to auto-load this command
# Council Help Display comprehensive help information for the LLM Council plugin. ## Overview The LLM Council is a multi-model deliberation system that consults three AI assistants (Claude, OpenAI Codex, and Google Gemini) to provide consensus-driven technical recommendations. ## Available Commands | Command | Description | |---------|-------------| | `/council "<question>"` | Summon the full council for deliberation | | `/council-help` | Display this help information | | `/council-status` | Check CLI availability and configuration | | `/council-config` | View and modify council settin...
/councilConvenes a multi-LLM council to provide advice, decision support, implementation plans, or gated implementation with configurable personas, depth, and providers.
/askQueries multiple AI agents (Gemini, OpenAI, Grok, Perplexity) for diverse perspectives on architecture decisions, technology choices, debugging, and security tradeoffs. Synthesizes recommendations with progress tracking.
/councilAssembles AI agent council with dynamic personas to generate parallel opinions on a query, anonymously rank them, and synthesize weighted final response.
/councilRuns a multi-LLM council with adversarial debate for a specified subagent (implementer, reviewer, architect, etc.) and task, producing specialized outputs like code, reviews, or plans.
/helpDisplays the Board of Advisors structure, council members, and a worked deliberation example in markdown.
/consensus-setupRuns interactive wizard to detect CLIs (kilo, codex, gemini, qwen), set OpenRouter key, and configure models/quorum for multi-model consensus reviews.
Share bugs, ideas, or general feedback.
Display comprehensive help information for the LLM Council plugin.
The LLM Council is a multi-model deliberation system that consults three AI assistants (Claude, OpenAI Codex, and Google Gemini) to provide consensus-driven technical recommendations.
| Command | Description |
|---|---|
/council "<question>" | Summon the full council for deliberation |
/council-help | Display this help information |
/council-status | Check CLI availability and configuration |
/council-config | View and modify council settings |
The council operates in three phases:
Each available council member independently analyzes your question and provides their response. Queries are sent in parallel for efficiency.
Each member reviews the responses from other members anonymously (labeled as "Response A" and "Response B"). This helps identify:
The Council Chairman (Claude Opus) synthesizes all opinions and reviews into a final verdict, including:
| Member | CLI | Role |
|---|---|---|
| Claude | claude | Required - Primary council member |
| OpenAI Codex | codex | Optional - Additional perspective |
| Google Gemini | gemini | Optional - Additional perspective |
# Technical architecture question
/council "What's the best approach for implementing real-time updates in a React application?"
# Code review
/council "Review this authentication implementation for security issues"
# Best practices
/council "How should I structure a microservices application for scalability?"
Use /council-config to:
Run /council-status to:
For installation instructions and detailed documentation, see the project repository.