PREVIEW. This is how it will look in an inbox; it was not sent to you. Each real email carries its own unsubscribe link and Keryn's postal address. KERYN HEADLINES Daily, Sunday 22 February 2026 Retrospective Covers 21 February 2026 00:00 to 22 February 2026 00:00 UTC Web: https://keryn.dev/headlines/daily/2026-02-22/ TOP DEVELOPMENTS [1] CISA Adds Two Actively Exploited Roundcube Flaws to KEV Catalog The Hacker News · Advisory · https://thehackernews.com/2026/02/cisa-adds-two-actively-exploited.html [2] When Backdoors Go Beyond Triggers: Semantic Drift in Diffusion Models Under Encoder Attacks arXiv · Research · https://arxiv.org/abs/2602.20193 [3] LoMime: Query-Efficient Membership Inference using Model Extraction in Label-Only Settings arXiv · Research · https://arxiv.org/abs/2602.18934 SAST [4] Automatic, Expressive, and Scalable Fuzzing with Stitching arXiv · Research · https://arxiv.org/abs/2602.18689 AGENTIC SDLC [5] Anthropic Launches Claude Code Security for AI-Powered Vulnerability Scanning The Hacker News · Announcement · https://thehackernews.com/2026/02/anthropic-launches-claude-code-security.html TRUST [6] From Docs to Descriptions: Smell-Aware Evaluation of MCP Server Descriptions arXiv · Research · https://arxiv.org/abs/2602.18914 [7] The coordination gap in frontier AI safety policies arXiv · Research · https://arxiv.org/abs/2603.10015 MODELS [8] MANATEE: Inference-Time Lightweight Diffusion Based Safety Defense for LLMs arXiv · Research · https://arxiv.org/abs/2602.18782 AI SECURITY [9] The Convergence of Schema-Guided Dialogue Systems and the Model Context Protocol arXiv · Research · https://arxiv.org/abs/2602.18764 ADVISORIES IN THIS WINDOW 3 PyPI advisories · 1 critical · 1 high · 1 moderate 2 Packagist advisories · 1 high · 1 moderate All advisories: https://keryn.dev/headlines/advisories/ Titles, links and short excerpts only; every item links to its source. Preview: the unsubscribe link and Keryn's postal address are added to each email when it is sent.