prompt

304 - Agent Skills - when to use them and why they matter

发表于 2026-02-09 hear injection skill

Agent Skills look simple, but they are one of the most powerful building blocks in modern AI coding workflows. In this episode, Kaushik and Iury break down when…

Mitigating prompt injection attacks with a layered defense strategy

发表于 2026-04-03 malicious instructions injection

（摘要暂无, 请查看原文）

Google Workspace’s continuous approach to mitigating indirect prompt injections

发表于 2026-04-03 defenses defense model

（摘要暂无, 请查看原文）

AI threats in the wild: The current state of prompt injections on the web

发表于 2026-04-24 ipi injections website

（摘要暂无, 请查看原文）

MFA prompt bombing (noun) [Word Notes]

发表于 2026-01-14 bombing glossary mfa

Please enjoy this encore of Word Notes. Hackers bypass, multifactor authentication schemes by sending a blizzard of spamming login attempts until the accounts o…

企业AI落地难？JVS-AI渐进式实践：安全可控、小切口验证、标准化沉淀与规模化复制

发表于 2026-11-12 prompt jvs ai

为什么90%的企业AI项目停在POC？——四大隐性障碍的技术归因多数企业AI试点无法跨越POC鸿沟，并非模型能力不足，而是落地路径存在结构性缺陷。作为一线开发者，我们观察到四个相互强化的底层障碍：数据主权模糊：客户信息、合同条款等敏感数据若经公有云API调用，面临出境合规与泄露失控风险，导致管理层不敢批准生产部署…