kelly-brandlisted

# Kelly Brand ## Overview Use this skill as the operator for a brand's **narrative single source of truth**. It keeps a file-backed App-in-Skill workbench over the whole message house — positioning, value pillars, the story bank, evidence-backed proof points, and vocabulary/tone guardrails — plus a drift monitor that flags off-brand usage across channels. The agent drafts; the human curates which drafts are **adopted as the canonical narrative**; the skill records those adoptions and can export the canonical narrative for downstream use. Everything is organized by Aaron's **TALE** discipline — **Trace → Architect → Land → Evaluate** — and every narrative asset carries the TALE `phase` and the `sub_skill` that produced it. Default interaction mode: App UI. Unless the user explicitly asks for chat-only handling, check onboarding/config, refresh or regenerate the local brand snapshot, start/reuse the local app with `app/start.sh`, and give the actual local URL. Use chat-only mode only when the user says "纯聊天", "chat only", "不要打开 UI", or similar; in that mode present numbered assets (`#1`, `#2`) and take verdicts in the conversation. Design philosophy follows the App-in-Skill specification paper: <https://mr-kelly.github.io/research/app-in-skill-specification-for-pairing-agent-skills-with-a-local-companion-ui.pdf>. The agent owns drafting and any external reads/writes; the app reads and writes local files only; the file handoff is the contract; the human's clicks are reserve