Use when you want a delegated second opinion or implementation from GPT (Codex), Gemini, Grok (xAI), or OpenRouter (config-driven, 400+ models) - seven expert subagents (Architect, Plan Reviewer, Scope Analyst, Code Reviewer, Security Analyst, Researcher, Debugger) and bundled ask-gpt/ask-gemini/ask-grok/ask-openrouter/ask-all/consensus commands, advisory (read-only) or implementation (write; Grok and OpenRouter are advisory-only).
Get Gemini second opinion on a question or current work. Single-shot, advisory, no contamination. Pinned to auto-gemini-3.
---
name: "ask-gemini"
description: "Ask Gemini for a single-shot second opinion."
---
<!-- GENERATED by scripts/sync-hosts.js - edit the source under prompts/, AGENTS.md, or examples/, then regenerate. -->
Call the `deliberation` MCP server's `ask-gemini` tool to ask Gemini for a single-shot second opinion.
Pass the full question and any needed context as the prompt - the expert does not share this session:
$ARGUMENTS
Read the tool result and apply your own judgment; do not relay raw output verbatim.Get GPT (Codex) second opinion on a question or current work. Single-shot, advisory, no contamination.
---
name: "ask-gpt"
description: "Ask GPT (Codex) for a single-shot second opinion."
---
<!-- GENERATED by scripts/sync-hosts.js - edit the source under prompts/, AGENTS.md, or examples/, then regenerate. -->
Call the `deliberation` MCP server's `ask-gpt` tool to ask GPT (Codex) for a single-shot second opinion.
Pass the full question and any needed context as the prompt - the expert does not share this session:
$ARGUMENTS
Read the tool result and apply your own judgment; do not relay raw output verbatim.Get Grok (xAI) second opinion on a question or current work. Single-shot, advisory, no contamination.
---
name: "ask-grok"
description: "Ask Grok (xAI) for a single-shot second opinion."
---
<!-- GENERATED by scripts/sync-hosts.js - edit the source under prompts/, AGENTS.md, or examples/, then regenerate. -->
Call the `deliberation` MCP server's `ask-grok` tool to ask Grok (xAI) for a single-shot second opinion.
Pass the full question and any needed context as the prompt - the expert does not share this session:
$ARGUMENTS
Read the tool result and apply your own judgment; do not relay raw output verbatim.Ask GPT, Gemini, Grok, and any configured OpenRouter models in parallel for independent second opinions, then synthesize and compare. Zero cross-contamination.
---
name: "ask-all"
description: "Ask GPT, Gemini, Grok, and OpenRouter in parallel for independent second opinions."
---
<!-- GENERATED by scripts/sync-hosts.js - edit the source under prompts/, AGENTS.md, or examples/, then regenerate. -->
Call the `deliberation` MCP server's `ask-all` tool to ask GPT, Gemini, Grok, and OpenRouter in parallel for independent second opinions.
Pass the full question and any needed context as the prompt - the expert does not share this session:
$ARGUMENTS
Read the tool result and apply your own judgment; do not relay raw output verbatim.Ask a single configured OpenRouter model for a second opinion. Advisory only. Single-shot or multi-turn.
---
name: "ask-openrouter"
description: "Ask a configured OpenRouter model for a single-shot second opinion."
---
<!-- GENERATED by scripts/sync-hosts.js - edit the source under prompts/, AGENTS.md, or examples/, then regenerate. -->
Call the `deliberation` MCP server's `ask-openrouter` tool to ask a configured OpenRouter model for a single-shot second opinion.
Pass the full question and any needed context as the prompt - the expert does not share this session:
$ARGUMENTS
Read the tool result and apply your own judgment; do not relay raw output verbatim.Arbiter-mediated consensus - GPT + Gemini + Grok (plus any configured OpenRouter delegates) review while Claude commits a blind verdict, adjudicates, and synthesizes. Converges only with cross-model agreement. Max 5 rounds.
---
name: "consensus"
description: "Fan out, then run one arbiter pass that synthesizes a single verdict."
---
<!-- GENERATED by scripts/sync-hosts.js - edit the source under prompts/, AGENTS.md, or examples/, then regenerate. -->
Call the `deliberation` MCP server's `consensus` tool to fan out, then run one arbiter pass that synthesizes a single verdict.
Pass the full question and any needed context as the prompt - the expert does not share this session:
$ARGUMENTS
Read the tool result and apply your own judgment; do not relay raw output verbatim.