An AI-Powered Natural Language Interface for Docker Container Health Monitoring: Architecture, Implementation, and Evaluation
Manual container management remains one of the more expensive and inconsistent stages of the DevOps lifecycle.Existing automated approaches often rely either on CLIbased tooling or on language-agnostic monitoring dashboards.This paper presents the design, implementation, and evaluation of an AI-powered Docker Natural Language Health Dashboard.The system accepts plain English commands from users, translates them into Docker operations using a Groq-hosted Large Language Model (llama-3.3-70b-versatile),and displays real-time container health metrics through a Streamlit web interface.The system analyzes commands using category-specific prompts covering container health, resource usage, lifecycle management, and system diagnostics.The LLM produces structured findings that are schema-validated, processed by a deterministic action engine, and executed against the Docker daemon with confidence and severity metadata.The system was evaluated on 20 natural language commands across 5 operational categories, achieving 100% precision, 100% recall, and 100% F1score with zero false-positive container actions.These results show that the proposed P Shenbagavalli, Dhivya R, Naveen M, Irfan K, Sasvin S K https://iaeme.com/Home/journal/FCSIT 2 editor
Authors
- Naveen M
- Dhivya R
- Sasvin S K
- P Shenbagavalli
- Irfan K
Publication Details
- Published
- 2026-11-13
- DOI
- https://doi.org/10.34218/fcsit_07_06_001
- Primary Topic
- Software System Performance and Reliability
- Type
- article
- Field-Weighted Citation Impact
- 0.00