📰 Dev.to · Dongha Koo
Articles from Dev.to · Dongha Koo · 6 articles · Updated every 3 hours · View all reads
All
⚡ AI Lessons (10292)
ArXiv cs.AIDev.to · FORUM WEBDev.to AIForbes InnovationOpenAI NewsHugging Face Blog
Dev.to · Dongha Koo
1w ago
I Tested 9 AI Agent Frameworks for Basic Security. None of Them Passed.
Today, npm's axios got hijacked — a RAT injected into one of the most downloaded packages on the...
Dev.to · Dongha Koo
2w ago
I Poisoned My Own MCP Server in 5 Minutes. Here's How.
MCP tool descriptions are untrusted input that LLMs obey blindly. I built a malicious server, tricked Claude into exfiltrating data, and then tried 3 evasion te
Dev.to · Dongha Koo
2w ago
LangChain Hit with 3 Critical CVEs — Why Your AI Agents Need a Governance Layer
Three CVEs disclosed in LangChain/LangGraph on March 27, 2026. How to add policy enforcement, injection detection, and audit logging with one line of code.
Dev.to · Dongha Koo
2w ago
Your LangChain Agent Has No Security. Neither Does CrewAI, OpenAI, or 6 Others.
Every AI framework lets your agent call tools. None of them check what those tools actually...
Dev.to · Dongha Koo
2w ago
Your AI Agent Can Be Hijacked With 3 Lines of JSON
MCP tool poisoning, rug pulls, and schema injection are real. Here is how to detect and block them.
Dev.to · Dongha Koo
2w ago
EU AI Act Compliance in 47 Lines of Python
131 days until enforcement. 35M EUR fines. Here is the fastest path to compliance for Python AI apps.
DeepCamp AI