MakerChecker scans AI agents for dangerous capabilities

MakerChecker is an open‑source tool hosted on GitHub. It enables developers to scan AI agents for potentially dangerous capabilities. The project provides a framework for automated safety checks. Users can define tests that probe for harmful outputs. The tool reports findings in a clear, actionable format. It is intended to help mitigate risks in AI deployments. Contributions from the community extend its coverage. The repository includes documentation on setup and usage.