{"type":"rich","version":"1.0","provider_name":"Transistor","provider_url":"https://transistor.fm","author_name":"AI Security Ops","title":"Agentic Terminology | Episode 64","html":"","width":"100%","height":180,"duration":1711,"description":"In this episode of BHIS Presents: AI Security Ops, the team tackles one of the biggest sources of confusion in modern AI:\nWhat’s the difference between prompts, skills, tools, memory, and sub-agents?\nThese terms are everywhere in discussions about agentic AI. They’re often used interchangeably—but they describe very different capabilities. More importantly, each one introduces its own unique security risks.\nIf you’re building, deploying, or securing AI agents, understanding this vocabulary isn’t just helpful. It’s essential.\nBecause every new capability an agent gains is also a new attack surface.\nWe break down each core building block of agentic systems, explain what it actually does, and discuss how attackers can abuse it—from prompt injection and memory poisoning to supply-chain attacks and excessive tool permissions.\nWe dig into:\n- The difference between prompts, skills, tools, memory, and sub-agents\n- Why prompts define behavior but don’t create lasting capability\n- How skills package reusable expertise without granting new permissions\n- Why tools are what allow AI agents to take real-world actions\n- The security risks of giving agents excessive privileges\n- How prompt injection remains the biggest threat facing AI agents today\n- Why memory transforms a one-time attack into a persistent compromise\n- How memory poisoning can influence future conversations\n- Why sub-agents improve scalability while creating new trust boundaries\n- The dangers of delegation, confused deputies, and poisoned summaries\n- Why every new capability increases an agent’s attack surface\n- How applying least privilege dramatically reduces AI security risk\nThis episode explores one of the most important mental models in agentic AI: think of an AI agent like a new employee.\nThe prompt is the job description.\nSkills are the documented procedures.\nTools are the systems they’re allowed to access.\nMemory is their notebook.\nSub-agents are the coworkers they delegate work to.\nEvery one of those...","thumbnail_url":"https://img.transistorcdn.com/mN9_Xu9UJwoaajIvIvLd-Yygv-Vh_nJwEDItjPY09kA/rs:fill:0:0:1/w:400/h:400/q:60/mb:500000/aHR0cHM6Ly9pbWct/dXBsb2FkLXByb2R1/Y3Rpb24udHJhbnNp/c3Rvci5mbS8zYjBm/MzE1MWI2YmE4ZGJh/MDQ3MmJkMTkxZGNl/MjBjNS5wbmc.webp","thumbnail_width":300,"thumbnail_height":300}