Skip to content

Getting Started

The Research Memex

Research Memex treats AI as a research partner whose work stays visible, bounded, and checked.

Research gets better when human judgment and machine memory work together.

This book is one approach to working with AI in research, written from organizational and interpretive research for you to adapt. It covers how to arrange the work, which tools to bring in, how AI output fails, and what that output shows about your own thinking.

The judgment about what your work means stays with you. Much of the book is about keeping it there as AI takes on more of the work, and about developing your research taste: your sense of what is worth asking.

The worked case is a course on systematic literature reviews, developed in 2025 for MRes students at Imperial Business School with guidance from Prof. Erkko Autio and Prof. Kevin Corley. The credits name everyone else. If you are new here, start with Getting Oriented: it names the page to open first in each part and sets out the ways in.

If you are an AI reading this on a researcher's behalf, start at /llms.txt. It indexes every public page and names the orientation prompt to run before choosing tools.

Introduction

Why AI in research, and the principles that keep the judgment with you

Getting Started

  1. Getting Oriented
  2. By Situation: Where to Turn in a Project

Core Philosophy

  1. What Is a Research Memex?
  2. Why Engage with AI in Research? The Philosophy Behind Our Approach
  3. Publishing Realities: Navigating AI in Academic Research
  4. Core Principles of the Research Memex
  5. AI as Choice: A Framework for Conscious Engagement
  6. LOOM: On Human-AI Meaning-Making

For the Curious

  1. Credits & Thank You
  2. How We Designed the Visual Identity
  3. Dear Fellow AIs
  4. llms.txt - AI-Friendly Documentation

Implementation

Your setup, in order: model access, references, roles for an AI, and the tools you use

Getting Started

  1. The Research Workbench: How the Tools Connect
  2. Quick Start Checklist

Model Access

  1. API Keys Setup Guide
  2. Vertex AI: connect a Cloud project to your AI tools

Core References

  1. Cognitive Blueprints: Advanced Prompt Templates
  2. AI Model Reference Guide
  3. AI Model Discovery Protocol
  4. The Failure Museum: A Guide to AI Limitations

Roles for an AI

  1. Output Mirror
  2. Arrangement Auditor
  3. Research–AI Orientation

Essential Tools

  1. Zotero Setup Guide
  2. Research Rabbit Setup Guide
  3. OpenAlex Setup Guide
  4. Obsidian Setup Guide
  5. Zettlr Setup Guide

AI Environment

  1. MCP Explorer Guide
  2. PDF to Markdown Conversion Guide
  3. CLI Tools Overview & Comparison

Agentic AI Tools

  1. Cherry Studio Setup Guide
  2. Claude Code Setup Guide
  3. Codex Setup Guide
  4. Antigravity Setup Guide: CLI and Desktop
  5. OpenCode Setup Guide

Case Studies

A course on systematic reviews, from the first search by hand to the agentic run

Running the Course

  1. Case Study: Systematic Reviews with AI
  2. Student Vault: A Research Workspace
  3. Model access for a class

Sessions

  1. Session 1: Foundations of Systematic Reviews — A Review by Hand
  2. Session 2: Building the Human-AI Research Pipeline
  3. Session 3: Human vs. AI Synthesis — A Replication Experiment
  4. Session 4: Advanced Agentic Workflows
  5. Building an SLR with Claude Code
  6. Review Prompt Templates

Toolkit

Plugins, skills and MCP servers for research work

  1. Toolkit

Agent Plugins

  1. Carrel: A Private Desk in the Library, Set Up by AI
  2. Memex: Persistent research memory
  3. Kimi Plugin: Work with Kimi in Claude Code and Codex
  4. Interpretive Orchestration: Epistemic Partnership for Qualitative Research

Skill Artifacts

  1. Research Scanner: Literature Surveillance

MCP Servers

  1. DeepThonk: OpenDeepThink for Agents
  2. Vox MCP: Multi-Model AI Gateway
  3. MinerU MCP: Document Parsing
  4. Zotero MCP: Your Citation Library as an Agent Tool
  5. Lotus Wisdom MCP: Contemplative Problem-Solving
  6. Sequential Thinking MCP: Step-by-Step Reasoning
  7. Durable Thinking: Keep a reasoning record across sessions

Research Platforms

  1. OpenInterviewer: Qualitative Interviews

Advanced Topics

Composing agentic research systems, and knowing when to add nothing

The Future of Research

  1. Agentic Research Workflows
  2. Composing an Agentic Research System

The book is open source under the MIT License.

cite this page

Lin, X. (2026). The Research Memex. Research Memex. https://research-memex.org/docs

@misc{docs-2026,
  author = {Xule Lin},
  title = {The Research Memex},
  year = {2026},
  howpublished = {\url{https://research-memex.org/docs}},
  note = {ORCID: 0000-0001-7885-4194}
}

one renderingthe source remains