In this paper we propose apirl, a fully automated deep reinforcement learning tool for testing rest apis It reads api documentation and learns which test inputs are most likely to reveal problems. The name april is derived from the latin word aperire, meaning to open. it is often associated with the spring season, symbolizing new beginnings, growth, and rebirth
As a name, april represents qualities such as freshness, vitality, and optimism. Think of apirl as a security tester that gets better at finding vulnerabilities the more it tests Apirl is a deep reinforcement learning tool to find bugs in rest apis
Definition of apirl in the definitions.net dictionary Information and translations of apirl in the most comprehensive dictionary definitions resource on the web. It was named tabularreinforcementlearning.jl at first johanni brea sent me the invitation to work on it together implemented several tabular rl related algorithms from the famous book reinforcement learning An introduction many deep rl related algorithms are implemented following that
Dqn related variants (relatively fast) policy gradient related variants (pretty fast) cfr related variants. This allows apirl to learn the subtleties relating to test outcomes, and generalise to unseen api endpoints This paper presents apirl, a system that uses this approach to test rest apis for security flaws
