📄️ Agent hooks
The committed lifecycle hooks only format successfully edited JavaScript/TypeScript files through installed oxfmt. Shared logic lives in scripts/agent-hooks/format.mjs; each native wrapper delegates to it.
📄️ Skills and Tools
Shared skills live in .agents/skills/. Edit these sources, then run yarn ai-workflow:sync to generate .claude/skills/ for Claude Code. Codex and Cursor discover .agents/skills/ directly; do not restore the duplicate .codex/skills/ or .cursor/skills/ roots.
📄️ Bug Investigation Workflow
Use this when a bug is reported in a specific file/line/code block.
📄️ Translations
The about site uses i18next JSON at about/public/translations//default.json. Docusaurus source translations live separately in docs/i18n/.
📄️ Commit and Issue Format
Use this when the user requests commit/issue wording or an authorized commit/issue is being created. Do not append suggestions to unrelated answers.
📄️ Long-running agent work
Use durable task state when work needs resumption or a handoff, or when a single run is long enough that context compaction could lose track of the remaining work. Small tasks do not need a board or progress file. For shared work, keep a concise feature-list.json and progress.md in a task-specific docs/agent-runs//, using existing templates where helpful.
📄️ Known Surprises
This file tracks repository-specific confusion points that caused agent mistakes.
🗃️ Templates
2 items