Research Methods
10 articles
Claude Code starter kit for economists
Seven starter resources for carrying project instructions and verification routines into later Claude Code sessions, with guarded installation steps.
A Pre-Analysis Plan for Your Coding Agent
A three-layer architecture, rule, gate, and verification, for keeping an AI coding agent disciplined, and why trained priors beat a system prompt.
What Agents Actually Do (And What They Don't)
A conceptual framework for specification-bounded research agents. Its linked timing example is article-reported, not a public workflow benchmark.
What We Mistake for AI Capability
A hypothesis about how task tolerance and specification shape perceived AI capability, with a controlled comparison still to run.
Research Phases Need Different Prompts
Exploration, implementation, and documentation require different AI prompting strategies. Match the prompt to the phase.
The Verification Tax: Every AI Output Needs Checking
Every AI output needs checking. Building verification into the research workflow to catch fabricated citations and errors before they compound.
Context Window Budgeting: Treating Tokens as a Finite Resource
Treating an AI context window as a finite budget: when to spawn sub-agents, when to work directly, and how to keep a long session from degrading.
End-of-Session Hygiene: What to Capture Before Context Resets
What to capture before an AI context window resets, and how five minutes of end-of-session notes saves twenty minutes rebuilding context tomorrow.
Antigravity Research Graphics Workflow
A practical workflow for using Antigravity image generation to explore a figure layout while keeping the final quantitative graphic reproducible in code.
The Cold Start Problem: Why the First Five Minutes Matter Most
Why the first five minutes of an AI coding session matter most, and how a CLAUDE.md context file solves the cold-start problem before work begins.