kelly-deal-scorerlisted

# Deal Scoring Desk ## Overview Use this skill as a local review-queue operator for a generic SME financing deal desk (private-credit / revenue-based-financing style). It ingests a mock queue of candidate businesses (name, category, city, monthly revenue history, requested principal, red flags) and computes a deterministic composite score (0-100) per candidate with a full, hand-recomputable breakdown: each sub-factor's raw score, weight, and contribution, plus a suggested revenue-share rate range. This is a generic, brand-free tool — it does not reference any specific real company, lender, or fund. **The scoring rubric is plain arithmetic in `lib/scoring.ts`, not an LLM or API call.** Every number the app shows can be recomputed with a calculator from the candidate's raw fields and the rubric weights in `config.json`. Default interaction mode: App UI. Unless the user explicitly asks for chat-only handling, check onboarding/config, seed or reuse the local batch, 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. ## App UI Screenshots <table> <tr> <td width="50%"><img src="assets/screenshots/overview.webp" alt="Deal Scoring Desk overview"></td> <td width="50%"><img src="assets/screenshots/candidate-detail.webp" alt="Deal Scoring Desk candidate detail"></td> </tr> <tr> <td><strong>Overview</strong><br>Queue-level summary header — score distribu