outlier-swarm

Orchestrates five parallel agents to produce contrarian research reports.

Updated Apr 10, 2026
One-click install
npx skills add https://github.com/GavinTPA/ai-powerhouse-starter --skill outlier-swarm
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: outlier-swarm
Source: https://github.com/GavinTPA/ai-powerhouse-starter/tree/main/my-skills/outlier-swarm
Command: npx skills add https://github.com/GavinTPA/ai-powerhouse-starter --skill outlier-swarm

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill finds unconventional, high-leverage approaches that mainstream research and coaching often miss by orchestrating focused, multi-angle investigations to surface validated contrarian insights.

Core Features & Use Cases

  • Five-agent orchestration: Runs Mainstream Mapper, Contrarian Hunter, Cross-Domain Thief, First Principles Assassin, and Failure Archaeologist in parallel to generate diverse perspectives.
  • Mission context integration: Loads a member's mission-context to tailor searches and prompts, with graceful fallback to generic framing.
  • Prior-knowledge and vault discovery: Attempts Smart Connections → Glob/Grep vault search before doing external web research.
  • Synthesis & Save: Produces a convergence-focused report highlighting overlaps of surprise and saves results to a vault or project research folder.
  • Use case: Use when you want to question conventional wisdom on pricing, onboarding, marketing channels, or product decisions and find evidence-backed alternatives.

Quick Start

Ask the assistant: Outlier swarm on [topic] to run the five-agent contrarian research and return an executive summary, convergence matrix, and action verdicts.

Frequently Asked Questions about outlier-swarm

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I find contrarian business strategies that mainstream research misses?

Finding overlooked strategies requires cross-domain synthesis and failure archaeology applied to your business context. This approach maps conventional wisdom, then actively hunts for validated contrarian alternatives and analogies from other industries to reveal unconventional, high-leverage approaches.

What is failure archaeology and how does it help with product or pricing decisions?

Failure archaeology analyzes past failures to uncover evidence-backed alternative strategies. By examining what went wrong in conventional approaches to pricing or onboarding, it reveals overlooked opportunities and helps you avoid repeating mainstream mistakes in your business decisions.

Can I run contrarian research using my local vault notes before searching the web?

Yes, contrarian research can use local vault notes through Smart Connections and Glob/Grep search. The system prioritizes discovering prior knowledge from your vault before falling back to external web search, ensuring your existing notes inform the multi-agent investigation.

How do I integrate mission context into cross-domain research for my coaching business?

Integrate mission context by loading your member's mission-context data to tailor searches and prompts. If unavailable, the system applies a generic fallback framing, ensuring the cross-domain research still aligns with your coaching business objectives.

Does multi-agent research synthesis work for small business marketing channel analysis?

Yes, multi-agent research synthesis works for small business marketing channel analysis by orchestrating parallel agents that map conventional channels and hunt contrarian alternatives. It generates a convergence report highlighting overlapping surprises and actionable verdicts for your business.

What are the limitations of using first-principles analysis for business strategy research?

First-principles analysis for business strategy research depends on mission-context integration; without it, the system uses generic framing. It also requires local vault discovery or web search capabilities to gather sufficient data, limiting effectiveness in offline environments.