IT Center Changes

First Major Release of the Production-Ready Product Following Completion of the Public Testing Phase

Features:

• Query Planner – RAG Retrieval Overhaul: A lightweight LLM call (gpt-5.4-mini) now runs before document retrieval to generate a cleaned, topic-focused search query, replacing the previous embedding-based topic-shift detection heuristic. Result: more accurate RAG retrieval and elimination of topic contamination (e.g., VPN and HPC being merged into a single incorrect query). Fully configurable at runtime through a new tab in the Env Manager, with a robust fallback ensuring that planner failures or timeouts never interrupt a chat session.
• Domain-Specific Search Language: HPC-/cluster-related queries are searched in English (matching the language of the knowledge base), while responses remain in the user’s preferred language.
• Explicit Language Switching: Instructions such as “answer in English” or “auf Deutsch bitte” now switch the response language reliably.
• More Accurate Incident Detection: Incident banners are displayed only when users explicitly ask about a disruption, rather than whenever a service is merely mentioned.
• Screenshot-Upload (Vision Support): Images can be attached via paste, file selection, or drag-and-drop. The client downsizes uploads to WebP, while the server validates and re-encodes image data using Pillow. Screenshots remain available throughout follow-up questions and are automatically removed after a configurable retention period.
• Dark Mode: New theme system with a dedicated dark stylesheet and theme toggle.
• Per-Message Test Cases and promptfoo Export: Administrators can mark individual question–answer turns as test cases, define the expected answer and correct documentation references, and export everything as promptfoo-compatible YAML files for automated quality evaluation.

Intern/Architektur

• The monolithic bot.py (775 lines) has been refactored into a structured app/bot/ package (pipeline, client, search, language, tokens, incidents, query_planner).
• routes.py has been modularized into subpackages under app/routes/.
• Integrated deployment webhook (seko_deploy_webhook), expanded test coverage, and CI/CD updates.

Bugfixes and Improvements:

• Source Attribution: Improved tracking of RAG source documents. Shared chats now display the referenced documentation under a dedicated Sources section.
• Feedback Review System Overhaul: Refactored review interface, introduced a flag queue for suspicious chats, and normalized the database schema using enums and lookup tables for feedback categories and AI models.
• Env Manager Redesign: Database-backed configuration management with bidirectional .env-synchronization, side-by-side configuration views, a dedicated write user, and export of merged .env files.
• Incident Processing: Improved prioritization by severity, time-based filtering, token-budget-aware truncation, and clickable incident links.
• Updated GPT Model Support: Token tracking and model routing updated for the latest GPT-5.x deployments.
• Evaluation API: New endpoint for automated evaluation runs.

Features:

• New Language Model:
  • Ritchy now uses GPT-5.2 by OpenAI as its underlying language model. Compared to the previous GPT-4.1 model, it provides improved reasoning capabilities and a deeper understanding of language.
• Automatic Integration of Incident Reports:
  • Ritchy can now automatically detect whether a user request relates to a current incident or maintenance activity of the IT Center. If a match is found, the corresponding incident report is directly incorporated into the response, including status information and a link to the incident page. Incidents are prioritized by severity (incident > partial incident > maintenance > warning > notice).
• Improved Document Search:
  • The way Ritchy accesses the IT Center documentation has been fundamentally redesigned. The search now uses a three-stage approach (keyword search, vector search, and semantic re-ranking) to retrieve more relevant articles. In addition, more document context is included in responses, resulting in more precise and complete answers.
• More Reliable Language Detection:
  • Automatic language detection has been made more robust. In particular, for queries containing English technical terms (e.g., product names like Cisco AnyConnect), the system now more reliably identifies the user’s actual language and consistently formulates the response in that language.
• Structural Improvements:
  • The internal architecture of the application has been modularized to simplify future extensions and maintenance. These changes do not result in any visible differences for end users.

• A chat routing is now set up during availability hours
• When selecting the support chat function on the IT Center website, on IT Center Help, or in RWTHmoodle, users can now choose whether they would like to chat with Ritchy or with IT ServiceDesk staff

Source: IT Center Help

• GPT-5 Chat integrated as the language model
  • GPT-5 Chat prioritizes conversational tone, immediate helpfulness, and faster responses
  • Optimized for clarity, brevity, friendliness, and consistent chat behavior

• Available via the footers of the various IT Center websites

• Ritchy is now accessible via IT Center Help
• Language Model: 4o Mini
• Public access outside regular business hours via Chat Support