We let an AI agent execute Bash and lived to talk about it — Sarah Sanders, PostHog
PostHog built Wizard, an AI tool that automates SDK installation and builds dashboards on developers' machines — approximately 8,000 users per week.
Sarah Sanders, a security engineer there, recognized that an AI system with command-line access is dangerous: an attack can originate from an insecure pull request that becomes part of the installation pipeline. Her audit found that even without obvious malice, two innocent functions together create gaps — attacks compose while code review does not. She built a detector based on YARA rules that reports suspicious patterns without acting itself, and layered an AI system on top that advises but never enforces.
Vibekollen prepared this summary with AI from the original publication. The content belongs to AI Engineer.
More to read
Lumaril, which develops AI software for optimizing industrial production by connecting to existing control systems, raised a $66M Series A at a $400M valuation
Techmeme 3 h ago
DeepSeek open sources Super-Agent as a research prototype, a proactive coding agent it says can monitor group chats for coding issues and suggest solutions
Techmeme 3 h ago
Insurance software maker Guidewire agrees to acquire Hyperexponential, which makes AI-powered insurance underwriting and pricing software, for $1.7B in cash
Techmeme 3 h ago
One word for AI agents in 2027
Nate Herk 7 h ago