REST API Testing with Verified LLM-Inferred Dependencies and Response-Driven Refinement
- Source
- arxiv.org
- Author
- Tu Nguyen, Thanh Nguyen, Huy Nguyen, Viet Nguyen, Tien N. Nguyen, Vu Nguyen
- Date

Context
APIPilot checks whether an AI's proposed relationships between API calls actually work before building tests around them. Some operations produce resources or values that later operations need. The system infers these dependencies from an API specification, tests them through real calls, then constructs workflows from the validated relationships. Responses also help it revise inputs and remove mistaken dependencies. The paper explains this execution-feedback loop.
Across 16 services, the authors report 92.3% operation coverage and an 88.1% workflow execution success rate. These measure different outcomes: reaching available operations and successfully running call sequences. The design makes execution evidence a prerequisite for trusting inferred dependencies.
- LLM — A large language model — the neural network behind tools like Claude and ChatGPT, trained on huge amounts of text to predict what comes next.
articleVibeCheck: Assessing the Quality of LLM-Generated Unit Tests: A Multi-agent Empirical Study across Heterogeneous RepositoriesAnika Tabassum, Mushahid Intesum, Md. Fahim Arefin, Tarannum Shaila Zaman
articleVerification as an Architectural Layer for LLM Agents: A V-Model Design, and a Pilot Study of Its Deterministic CoreAli Afoud, Jie JW Wu
articleHxAgent: Iterative Agent Planning for End-to-End Web Application TestingTu Nguyen, Duy Cao, Viet Nguyen, Phu Nguyen, Vy Le, Nguyen TK Nguyen, Tien N. Nguyen, Vu Nguyen
Checking sign-in…
Loading comments…