https://bluesquadron.dev/tags/agentic-security/Recent content in Agentic Security on BlueSquadron — Security engineering for AI systemsHugo -- gohugo.ioen-usflorent.batard@gmail.com (Florent Batard)florent.batard@gmail.com (Florent Batard)Mon, 31 Aug 2026 00:00:00 +0000https://bluesquadron.dev/posts/securing-the-ai-driven-development-lifecycle/Mon, 31 Aug 2026 00:00:00 +0000florent.batard@gmail.com (Florent Batard)https://bluesquadron.dev/posts/securing-the-ai-driven-development-lifecycle/AI-DLC gives you the strongest process integrity I have seen in any methodology — approval gates, audit shards, element-level traceability. None of it tells you the code is safe. Those are different claims, and closing the gap phase by phase is the work.https://bluesquadron.dev/posts/security-coach-for-ai-agents/Tue, 25 Aug 2026 00:00:00 +0000florent.batard@gmail.com (Florent Batard)https://bluesquadron.dev/posts/security-coach-for-ai-agents/Every team debating whether to trust the model has already made the decision that matters: where the boundary lives. A prompt is a promise. A tool interface is a constraint. Four patterns for telling them apart.https://bluesquadron.dev/posts/agent-driven-security-operations/Wed, 08 Jul 2026 00:00:00 +0000florent.batard@gmail.com (Florent Batard)https://bluesquadron.dev/posts/agent-driven-security-operations/Tuning detections until the volume fits your headcount means choosing your visibility based on your staffing. The fix is not more analysts — it is separating the three layers of triage by what each one is allowed to be wrong about.