Microsoft Defender email security benchmarking: Key insights from one year of data

2026-06-17T14:52:18Z9213187f2b1ff399e8140553c19527fc6f13712357a8d529dba529a505f7e908
AI-securityASSERTAnthropicAzure AICI/CDClaude CodeGitHub ActionsM365 CopilotMDASHagentic-AIcredential-stealingdependency-confusionemail-securityendpoint-protectionincident-responsenpmprompt-injectionred-teamingsocial-engineeringsupply-chaintelemetry

What happened

Collection of Microsoft Security Blog posts (Jun 2026) covering multiple high‑risk supply‑chain and AI/agent security issues. Key items: a large-scale npm supply‑chain campaign (“Miasma”) that compromised >90 @redhat-cloud‑services package versions to steal credentials and propagate; a dependency‑confusion campaign using 33 malicious npm packages to profile developer/build environments; a prompt‑injection pathway in Anthropic’s Claude Code GitHub Action enabling potential workflow secret access and guidance for securing AI‑powered CI/CD; updated taxonomy of seven new failure modes in agentic A

Why it matters

A reviewed impact interpretation has not been published for this record.

Evidence and limitations

Source ID
microsoft_security_blog
Record identifier
9213187f2b1ff399e8140553c19527fc6f13712357a8d529dba529a505f7e908
Enrichment time
2026-06-17T14:52:18Z
AI-assisted enrichment
Yes

This record may overlap with other records. Its enrichment can be incomplete or wrong, and machine assistance was used. Validate consequential decisions against the linked source and your own environment.