Research pipeline that decides depth after the evidence — source-grade weighted scoring, worker-hidden thresholds, and type-driven investigation frames.
1 file for Claude Code and Codex: deep-research CLAUDE.md — 2,725 tokens loaded in every session.
CLAUDE.md B 2,725 tok These files are treylom/deep-research's own configuration — they tell Claude Code and Codex 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.