council
Send one prompt to multiple AI backends in parallel and gather all responses for synthesis. Helps in architectural decisions and controversial calls by surfacing diverse perspectives.
Instructions
Pose one prompt to several AI backends in parallel and return all of their responses for Claude to synthesize. Backend selection is driven by topic (e.g. coding routes to qwen + local, reasoning routes to deepseek). confidence_needed controls how many backends are queried — high (4), medium (3), low (2). Use for architectural trade-offs, controversial calls, or anywhere dissent surfaced cheaply (~1-2s for 2-3 backends) is more useful than a single answer. For a single backend query, use ask. Read-only: makes N parallel HTTP calls; never writes to disk. Returns: {success, topic, strategy, confidence_needed, backends_queried:[names], backends_responded:[names of those that succeeded], responses:[{backend, success, content, response_time, error?}], processing_time_ms, metrics, synthesis_hint (suggestion to Claude on how to synthesize)}.
Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| topic | Yes | Topic category - determines which backends are consulted: coding (nvidia_qwen, local), reasoning (nvidia_deepseek), architecture (nvidia_deepseek, nvidia_qwen), general (gemini, groq), creative (gemini, nvidia_qwen), security (nvidia_deepseek, nvidia_qwen), performance (nvidia_deepseek, local) | |
| prompt | Yes | The question or topic for the council to deliberate on | |
| max_tokens | No | Maximum tokens per backend response | |
| num_backends | No | Override number of backends to query (optional - auto-calculated from confidence_needed) | |
| confidence_needed | No | Required confidence level - determines number of backends: high (4 backends), medium (3 backends), low (2 backends) | medium |