agora-council
by robbyczgw-cla
Web & Frontend DevelopmentBrief
Multi-agent debate council — spawns 3 specialized sub-agents in parallel (Scholar, Engineer, Muse) to tackle.
Description
What This Skill Does
Agora spawns three specialized sub-agents in parallel (Scholar, Engineer, and Muse) to analyze a problem from research, logic, and creative angles simultaneously. A main Captain agent decomposes the query, dispatches to each specialist, and synthesizes a final answer. Model selection per role is configurable via flags or presets.
Running three specialized models in parallel covers blind spots that a single-model query misses, since each role brings a different reasoning style to the same problem.
When to Use It
- Deciding between competing technical architectures
- Researching and fact-checking a topic before publishing
- Evaluating a business or product strategy from multiple angles
- Debugging a complex system with logical and creative input
- Getting a balanced take on a controversial technical decision
Example Workflow
User asks: /agora Should I use PostgreSQL or MongoDB for a new SaaS app?
- 1
Captain parses the query and decomposes it into sub-tasks for Scholar, Engineer, and Muse
- 2
Scholar searches for recent benchmarks, case studies, and community recommendations
- 3
Engineer analyzes schema requirements, query patterns, and scaling trade-offs
- 4
Muse considers developer experience, team familiarity, and long-term maintainability
- 5
Captain synthesizes all three responses into a final recommendation with confidence level and dissent notes
A council answer recommending PostgreSQL for structured SaaS data with ACID requirements, noting MongoDB's flexibility as a dissenting angle for rapid prototyping phases
Installation
clawhub install robbyczgw-cla/agora-councilRun the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.