Establishes the global AI development config repo from scratch: - Chunk 1: repo skeleton, install.sh, statusline, deploy manifest - Chunk 2: core instructions (coding/git/testing), CLAUDE.md rewrite (always-on + content index two-tier model), docs restructure, 6 ADRs, ROADMAP.md, .gitkeep placeholders - Bootstrap skills in .claude/skills/ (to be catalogued and migrated to .agents/skills/ in Chunk 3) Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
679 B
679 B
Two-tier CLAUDE.md: always-on rules + on-demand content index
The deployed ~/.claude/CLAUDE.md has two sections: (1) a short always-on section with universal rules loaded every session, and (2) a content index of pointers to on-demand files the agent reads when the task warrants it. The alternative — a single large file with all rules, workflows, and conventions — would load coding standards, git rules, and testing conventions into every session, including sessions that never touch code or run tests. Context is a scarce resource. Keeping the always-on section under 30 lines ensures it costs almost nothing; the agent pulls deeper content only when it's relevant.