research

Execute multi-phase USACF research investigations with parallel discovery and adversarial review.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/ste-bah/archon --skill research-ste-bah
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: research
Source: https://github.com/ste-bah/archon/tree/main/.claude/skills/research
Command: npx skills add https://github.com/ste-bah/archon --skill research-ste-bah

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This capability provides a disciplined, agent-based framework to conduct deep, structured research using the USACF framework, turning complex topics into organized, repeatable analyses.

Core Features & Use Cases

  • Phase-based workflow: Pre-search meta-analysis, discovery, gap/risk analysis, and synthesis phases orchestrated by multiple agents.
  • Parallel exploration: Executes discovery and synthesis tasks in parallel to accelerate insights.
  • Adversarial review: Built-in red-teaming to challenge findings and improve reliability.
  • Deliverables & memory: Stores outputs in a documented docs/research structure for traceability and reuse.
  • Topic-agnostic: Applicable across domains such as technology, policy, and science.

Quick Start

Provide a research topic and run the USACF pipeline to generate a phased report.

Frequently Asked Questions about research

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

FAQPage Schema
How do I conduct multi-agent research on complex topics using AI?

Multi-agent research uses the USACF framework to generate super-prompts and execute phase-based investigations across technology, policy, or science domains. It orchestrates parallel discovery and synthesis to accelerate deep, structured analysis.

What is adversarial review in AI research workflows?

Adversarial review is a built-in red-teaming phase that challenges research findings to improve reliability. It operates within multi-phase investigations to test conclusions, identify gaps, and synthesize more robust final reports.

Can I use this multi-agent research framework for policy and science topics?

Yes, the USACF research pipeline is topic-agnostic and applies across domains like technology, policy, and science. It handles complex topics requiring pre-search meta-analysis, targeted discovery, and adversarial synthesis.

How do I start a structured research investigation using the USACF framework?

Provide a research topic and run the USACF pipeline to generate a phased report. The workflow executes pre-search meta-analysis, parallel discovery, gap analysis, and synthesis, storing all deliverables under a documented docs/research structure.

Does the multi-agent research pipeline require web and code search tools?

Yes, executing parallel agent orchestration and multi-phase investigations requires access to web and code search tools. The framework also needs USACF documentation and memory storage to maintain traceability and reuse outputs.

What limitations exist when running parallel discovery and synthesis agents?

This approach requires USACF documentation, parallel agent orchestration capabilities, and active web or code search tools. Without these dependencies, the phase-based discovery and adversarial review processes cannot execute properly.