prism

Orchestrate multi-session research audits across multiple LLM vendors.

5|Updated Apr 23, 2026
One-click install
npx skills add https://github.com/Ronkupper/PRISM --skill prism-ronkupper
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: prism
Source: https://github.com/Ronkupper/PRISM/tree/main
Command: npx skills add https://github.com/Ronkupper/PRISM --skill prism-ronkupper

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

PRISM solves the challenge of conducting rigorous, multi-session research and audits by breaking complex problems into atomic, specialist prompts and converging them into a single, living Master document.

Core Features & Use Cases

  • Multi-Vendor Orchestration: Dispatches research tasks to the best-suited LLM (Claude, ChatGPT, Gemini, Perplexity) for each specific sub-task.
  • Continuous State Management: Maintains engagement state across sessions using a Master document and a What's next artifact, ensuring no work is lost.
  • Lens Library Integration: Uses a structured catalog of audit scopes to grade strategies and surface gaps before execution.

Quick Start

Run a full PRISM audit on your subject by invoking the skill and stating your research topic in plain language.

Frequently Asked Questions about prism

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

FAQPage Schema
How do I orchestrate multi-session research across different LLM vendors?

You can orchestrate multi-session research by decomposing complex problems into atomic prompts for multi-vendor LLM execution. This framework dispatches tasks to the best-suited models and converges outputs into a persistent Master document.

What is the best way to maintain research state across multiple LLM sessions?

Maintaining research state across sessions requires a persistent Master document and iterative setup probes. This framework manages engagement state continuously to ensure no work is lost and scope remains complete.

Can I use this framework to audit complex strategies using multiple LLMs?

Yes, you can audit complex strategies by utilizing a structured catalog of audit scopes. The framework grades strategies and surfaces gaps before execution, ensuring rigor across diverse model outputs.

How do you ensure convergence and rigor when aggregating outputs from diverse LLMs?

Ensuring convergence and rigor when aggregating diverse LLM outputs relies on a telemetric context-pressure framework. This mechanism maintains analytical focus and prevents scope drift across multi-vendor execution.

Do I need specific dependencies or scripts to run multi-vendor LLM audits?

No specific external dependencies are required to run multi-vendor LLM audits. The framework operates using internal scripts to manage orchestration, state, and convergence for your research workflows.

When should I not use an LLM-orchestrated framework for research analysis?

You should not use an LLM-orchestrated research framework when your problem scope is atomic and requires a single model session. It is designed for complex problems needing decomposition and multi-session state management.