Guard Research

Reproducible research on AI coding agent runtime security. All datasets are downloadable and the methodology is documented.

• AI Coding Agent Runtime Security Benchmark

  A reproducible benchmark comparing runtime security controls across five AI coding agent harnesses. 11 scenarios, 4 comparators, 220 results.

  Published 2026-07-26

• Benchmark Methodology

  How the benchmark is structured: scenarios, comparators, metrics, fixture design, and limitations.

  Published 2026-07-26