Python multi-agent board-records assistant with a governance agent that judges every answer against an externalized policy before release — allow, revise, or block. Anonymity is enforced by the schema, not the prompt. MCP boundary with audit trail.
2 files for Claude Code: redteam-anonymity, boardroom-agents CLAUDE.md — 2,969 tokens loaded in every session.
CLAUDE.md A 2,969 tok .claude/skills/redteam-anonymity/SKILL.md A 75 tok These files are edgemoorlf/boardroom-agents'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.