Citizens often face difficulty locating clear information about government services, welfare schemes, scholarships, eligibility conditions and required documents because the information is distributed across multiple government portals. This paper presents CivicGuide AI, a webbased Indian civil-services information portal with a bilingual citizen-assistance chatbot. The running application is implemented using Python and FastAPI. Its chat endpoint accepts a user message and language preference, then passes the query to a dedicated processing engine. The engine uses text normalization, keyword and phrase matching, intent-specific conditions and curated structured datasets to generate responses about scholarships, caste certificates, voter services, civic services, examinations, welfare schemes and state-specific information. The portal also provides dedicated information pages, search and an eligibility-checking feature. The implementation examined here is rule-based and data-driven; it does not use the separate RAG pipeline, ChromaDB, sentence embeddings or Cohere generation service present elsewhere in the repository. This paper describes the application launched by the project entry point, its architecture, query-processing workflow, functionality and limitations. Initial manual functional checks using multiple questions, including specific servicerelated queries, produced relevant responses in the cases tested. A larger controlled evaluation with a documented test set and measured response times remains necessary

Maseeha Banu, C R Pruthvik, Teju R S, Teju R S, Deekshitha K P, Mohammed Arshil, "CivicGuide AI: An Intelligent Government Services Information Assistant Using Rule-Based Query Processing" International Journal of Computer Science and Engineering Development, V1(4): Page(240-242) September - October 2026. ISSN: 3139-0862. www.ijcsed.com. Published by Scientific and Academic Research Publishing.