What problem does it solve? Caseworkers at the Chamber of Crafts must manually check every application for the Hessen advancement grant (Meisterprämie) against ten legal criteria, verify a three-month deadline, and extract personal and payment data from scanned documents — a slow, error-prone process. This Skill automates the review and prepares a documented recommendation while leaving the final decision to a human. ## Core Features & Use Cases - Ten-criteria eligibility check (K1–K10): Evaluates DQR level, exclusion criteria, complete examination certificate, the three-month filing deadline, Hessen residence/employment connection, mandatory form fields, and signature — each rated fulfilled / not fulfilled / unclear with source citations. - Traffic-light recommendation: Outputs GRUEN (approval), GELB (follow-up request or clarification), or ROT (rejection proposal) with the decisive reason, plus a structured JSON case file and a draft response letter for human approval. - Dual intake paths: Processes documents uploaded directly to the chat or retrieves applications from the email mailbox, classifying each attachment (e.g., full master certificate vs. partial result notice vs. Aufstiegs-BAföG forms). - Use Case: A clerk drops a scanned application and certificate into the chat, states the receipt date, and receives a complete audit-ready review report with masked personal data and a draft letter — without anything being sent automatically. ## Quick Start Upload the application documents to the chat, enable agent mode with a vision-capable model, and type /aufstiegspraemie together with the date the application was received by the chamber.
