roundtable - Multi-Agent Debate Council
Spawns Scholar, Engineer, and Muse agents in parallel, optionally runs cross-examination, and synthesizes their findings into a final response.
Tags
Updated: 2026-10-07Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Spawn specialized sub-agents
- Coordinate parallel Round 1
- Run optional cross-examination
- Synthesize findings and dissent
- Configure models and templates
- Estimate and control budget
- Save council session logs
Inputs
- User question
- Command flags
- Model configuration
- Role templates
- Existing config.json
- Web search results
Outputs
- Council responses
- Cost estimates
- Setup prompts
- config.json
- Session log files
Requirements
- Parallel sub-agent support
- Configured language models
- Web search access
- Workspace write access