LLM Action 제어 계층 도입을 통한 Discount Policy 준수 및 CRM 거버넌스 확보
We rolled out Copilot. Sales asked it for a 25% discount. Policy allows ten.
We rolled out Copilot. Sales asked it for a 25% discount. Policy allows ten.
Making Human Approvals Trustworthy
I gave Claude SSH access to my server — here's the consent gate that makes it safe
An AI agent acted across two companies. Whose audit log knows which human?
An AI agent exported a patient record. Your logs can't say who told it to.
사전 승인 없는 LPR 조회 권한으로 인한 개인정보 오용 및 설계 결함 분석
I Almost Hashed a Document Number That Needed to Be Read Again
Why We Don't Call Cosmic an "AI CMS"
coding agents made repositories the security boundary
I Let AI Agents Attack My Permission Gateway for a Week. Here's What Broke.
I got tired of writing the same history table boilerplate, so I built a Postgres extension
I added real-time activity logging and security scoring to my Claude Code dashboard
Building an Event Sourced Video Moderation Audit Log with PHP and SQLite
AI Agent Workboards Need Audit Controls Before They Need More Agents
NotebookLM Automation With notebooklm-py: Useful, But Classify Data First
Chatbots GPT et conformité au RGPD : comment les entreprises françaises abordent l’adoption de l’IA
Putting my own MCP server behind my own MCP gateway
I Built an Oracle DBA That Lives in Telegram. It Cut a 500K-Row Scan to 5 - After Asking Permission.
I vibe-coded a Postgres workspace because pgAdmin and DBeaver don't fit how I work
Desktop Agents Are The Next Big Trust Problem