Prompt injection in HTTP headers: what 217 million web responses show
Summary
A detailed security-focused study on prompt injection via HTTP headers observed across 217 million web responses. It analyzes how headers can direct AI agents, the prevalence of AI-targeted headers, and practical defenses for securing AI fetch flows. Useful for teams building AI-enabled workflows to understand risks and implement governance.