Aurora ASR Self-Healing Agent
What it is
Aurora ASR is an automated software repair system that uses Reinforcement Learning and LLMs to detect and fix bugs in code. It operates within a sandboxed environment, utilizing a 3-tier task system ranging from simple typos to complex logic errors, and supports both real LLM inference and a mock mode for offline testing.
The problem it solves
Manual debugging is time-consuming and error-prone; this system aims to automate the detection and repair of software bugs.
What you could use it for
- Automated debugging of Python code in CI/CD pipelines
- Training reinforcement learning agents on software repair tasks
- Demonstrating LLM-based code correction capabilities in a safe sandbox
Read the source ↗Back to the library →
Summary written by an AI model from the source; check the source before you rely on it.