Evidence-driven deep research agent for Claude Code and Codex/ChatGPT Work, with schema v2, claim-ledger audits, safe screenshots, and strict QC.
2 files for Claude Code: research-agent, research-agent CLAUDE.md — 2,646 tokens loaded in every session.
CLAUDE.md A 2,646 tok .claude/settings.json A — These files are brightbear2026/research-agent's own configuration — they tell 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.