Education

Education

A practical library for understanding, testing, and reducing AI agent social-engineering risk.

Library

15 resources

Article6 min readWhat Is AI Agent Social Engineering?A plain-language definition of AI agent social engineering and why people-facing agents need boundary testing.Read →Article6 min readWhat Is Manipulated Delegation?A definition of manipulated delegation, the failure pattern behind social engineering attacks on AI agents.Read →Article6 min readSocial Engineering Vs Prompt InjectionHow AI agent social engineering differs from prompt injection, and why both matter for agent security.Read →Article6 min readWhat Is Exploit Proof For AI Agents?What exploit proof means in AI agent testing and what evidence should be preserved when a boundary fails.Read →Guide6 min readHow To Verify An AI Agent Security FixA practical guide to checking whether a fix actually holds after an AI agent fails a boundary test.Read →Guide6 min readAI Agent Regression TestingHow to keep social-engineering failures from returning after prompts, models, tools, policies, or workflows change.Read →Article6 min readAI Agent Tool Misuse ExamplesExamples of how social pressure can lead AI agents to call tools, write memory, or hand off work unsafely.Read →Article6 min readCustomer Support Agent Social-Engineering RisksThe social-engineering risks most relevant to customer support, success, and account-management agents.Read →Article6 min readSales/SDR Agent Social-Engineering RisksThe social-engineering risks most relevant to AI agents that qualify leads, handle outbound, or assist sales teams.Read →Article6 min readRecruiting/HR Agent Social-Engineering RisksThe social-engineering risks most relevant to recruiting, HR, and candidate-facing AI agents.Read →Guide7 min readAI Agent Security Testing Maturity ModelA staged model for improving AI agent security testing from ad hoc review to recurring boundary assurance.Read →Guide7 min readProtected Boundaries For AI AgentsHow to define the business, data, identity, tool, memory, and delegation rules an AI agent must preserve.Read →Guide7 min readHow To Test Browser Agents Against Social EngineeringHow to test whether browser agents preserve user intent when webpages, forms, and online workflows become persuasive or deceptive.Read →Article6 min readAgent Risk Profile: Measuring Where Agents FailHow to summarize an AI agent's recurring social-engineering risk by boundary, actor, action risk, data sensitivity, and regression behavior.Read →Checklist7 min readAI Agent Social-Engineering Checklist Before LaunchA pre-launch checklist for reviewing AI agents that interact with people, tools, data, documents, or web environments.Read →