Turn specs into evals for any agent with ASSERT

A set of Microsoft Security Blog posts covering AI- and developer-focused threats and defenses. Highlights include: ASSERT, an open-source framework to convert natural-language specs into executable evaluations; guidance and a playbook for reconstructing AI activity in Microsoft 365 Copilot and Azure AI; analysis of threat actors leveraging AI branding for social engineering; discovery of a prompt-injection pathway in the Claude Code GitHub Action that could expose workflow secrets (responsible disclosure and mitigation by Anthropic); an updated taxonomy of failure modes for agentic AI basedon