AI-moderated user-research pipeline: voice/text interviews → auto-analysis → semantic search with cited quotes, exposed as an MCP server. Next.js 15 + Turso + OpenRouter, with an eval harness.
2 files for Codex, OpenCode and Claude Code: research-loop AGENTS.md, research-loop CLAUDE.md — 1,281 tokens loaded in every session.
AGENTS.md A 1,180 tok CLAUDE.md A 101 tok These files are JerryWhites/research-loop's own configuration — they tell Codex, OpenCode and Claude Code how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.