--- title: "#promptinjections articles | Hashnode" description: "Explore 79 articles tagged #promptinjections on Hashnode. Tutorials, guides, and how-tos from the community." url: "https://hashnode.com/tag/promptinjections" --- Tag # #promptinjections 79 posts · 2 followers [Jenna's Runbooks](https://hashnode.com/@jennasrunbooks) · [4d ago](https://hashnode.com/posts/defending-an-mcp-server-against-prompt-injection-through-cloudtrail/6aefce9e8620ee70ad53ee89) ### [Defending an MCP server against prompt injection through CloudTrail](https://blog.jennasrunbooks.com/defending-an-mcp-server-against-prompt-injection-through-cloudtrail?utm_source=hashnode&utm_medium=feed) Microsoft's spotlighting defense in its simplest form, and the layers that make it enough 5 min read · [blog.jennasrunbooks.com](https://blog.jennasrunbooks.com/defending-an-mcp-server-against-prompt-injection-through-cloudtrail?utm_source=hashnode&utm_medium=feed) [1](https://hashnode.com/posts/defending-an-mcp-server-against-prompt-injection-through-cloudtrail/6aefce9e8620ee70ad53ee89) [Fonz](https://hashnode.com/@aiza-hextyx) · [Oct 23](https://hashnode.com/posts/how-would-you-detect-and-mitigate-adversarial-prompt-injection-attacks-against-autonomous-ai-agents/6adbd28b15e272f742f79c5e) ### [How Would You Detect and Mitigate Adversarial Prompt Injection Attacks Against Autonomous AI Agents?](https://aiza-hextyx.hashnode.dev/how-would-you-detect-and-mitigate-adversarial-prompt-injection-attacks-against-autonomous-ai-agents?utm_source=hashnode&utm_medium=feed) a serious security challenge: what happens when an AI agent follows malicious instructions hidden inside the data it is supposed to process? 13 min read · [aiza-hextyx.hashnode.dev](https://aiza-hextyx.hashnode.dev/how-would-you-detect-and-mitigate-adversarial-prompt-injection-attacks-against-autonomous-ai-agents?utm_source=hashnode&utm_medium=feed) [1](https://hashnode.com/posts/how-would-you-detect-and-mitigate-adversarial-prompt-injection-attacks-against-autonomous-ai-agents/6adbd28b15e272f742f79c5e) [Gil Neto](https://hashnode.com/@gilneto8) · [Oct 22](https://hashnode.com/posts/a-bank-statement-can-carry-hidden-orders-for-an-ai/6ad9ef00ac4e0bd117bf8a6c) ### [A bank statement can carry hidden orders for an AI](https://blog.gil-neto.com/a-bank-statement-can-carry-hidden-orders-for-an-ai?utm_source=hashnode&utm_medium=feed) WhatsApp's help pages had text only a chatbot could read. I planted the same kind of text in a bank statement and ran it through my extractor. The money held, the hidden order didn't get stopped. 7 min read · [blog.gil-neto.com](https://blog.gil-neto.com/a-bank-statement-can-carry-hidden-orders-for-an-ai?utm_source=hashnode&utm_medium=feed) [Nagappan S](https://hashnode.com/@imkarthik03) · [Oct 9](https://hashnode.com/posts/prompt-injection-why-ai-agents-shouldn-t-trust-their-own-tools/6ac8900f47ebf6b7c629c6ca) ### [Prompt Injection: Why AI Agents Shouldn't Trust Their Own Tools](https://nagspidey.hashnode.dev/prompt-injection-why-ai-agents-shouldn-t-trust-their-own-tools?utm_source=hashnode&utm_medium=feed) Prompt Injection 9 min read · [nagspidey.hashnode.dev](https://nagspidey.hashnode.dev/prompt-injection-why-ai-agents-shouldn-t-trust-their-own-tools?utm_source=hashnode&utm_medium=feed) [4](https://hashnode.com/posts/prompt-injection-why-ai-agents-shouldn-t-trust-their-own-tools/6ac8900f47ebf6b7c629c6ca) [Christian Johannsen](https://hashnode.com/@cjohannsen) · [Sep 10](https://hashnode.com/posts/permissions-vs-policies-in-llm-systems-what-matters-where/6aa2d6ca02629ff0ef91bd19) ### [Permissions vs. Policies in LLM Systems: What Matters Where](https://breakglass.hashnode.dev/permissions-vs-policies-in-llm-systems-what-matters-where?utm_source=hashnode&utm_medium=feed) Permissions set what an LLM agent can do. Policies decide what it should do. Why system prompts aren't security controls, and where each belongs. 7 min read · [breakglass.hashnode.dev](https://breakglass.hashnode.dev/permissions-vs-policies-in-llm-systems-what-matters-where?utm_source=hashnode&utm_medium=feed) [Sravan Vidiyala](https://hashnode.com/@sravanvidiyala) · [Aug 10](https://hashnode.com/posts/my-test-suite-was-green-my-detector-was-wrong/6a796091a1f0931c50030008) ### [My test suite was green. My detector was wrong.](https://blog.sravanvidiyala.com/my-test-suite-was-green-my-detector-was-wrong?utm_source=hashnode&utm_medium=feed) I built a scanner to catch poisoned MCP tool descriptions, pointed it at 13,474 real ones, and watched it flag a tool called prompt\_injection\_scan as prompt injection. 9 min read · [blog.sravanvidiyala.com](https://blog.sravanvidiyala.com/my-test-suite-was-green-my-detector-was-wrong?utm_source=hashnode&utm_medium=feed) [3](https://hashnode.com/posts/my-test-suite-was-green-my-detector-was-wrong/6a796091a1f0931c50030008) [Ayesha Khan](https://hashnode.com/@ayeshakhalil) · [Aug 5](https://hashnode.com/posts/i-built-a-vulnerable-llm-bot-and-caught-my-own-judge-lying-to-me/6a72e4b4e2fe41e4c579aa5f) ### [I Built a Vulnerable LLM Bot and Caught My Own Judge Lying to Me](https://ayeshakoder.hashnode.dev/i-built-a-vulnerable-llm-bot-and-caught-my-own-judge-lying-to-me?utm_source=hashnode&utm_medium=feed) I built a vulnerable LLM bot, attacked it, then caught my own output-validation judge hallucinating a false safety verdict. 8 min read · [ayeshakoder.hashnode.dev](https://ayeshakoder.hashnode.dev/i-built-a-vulnerable-llm-bot-and-caught-my-own-judge-lying-to-me?utm_source=hashnode&utm_medium=feed) [1](https://hashnode.com/posts/i-built-a-vulnerable-llm-bot-and-caught-my-own-judge-lying-to-me/6a72e4b4e2fe41e4c579aa5f) [tiago](https://hashnode.com/@notthq) · [Jul 24](https://hashnode.com/posts/i-tried-to-hack-my-own-ai-agent-i-failed-that-was-the-best-result-i-could-have-asked-for/6a6352ed8cf7a5160d499027) ### [I tried to hack my own AI agent. I failed. That was the best result I could have asked for.](https://nott.hashnode.dev/i-tried-to-hack-my-own-ai-agent-i-failed-that-was-the-best-result-i-could-have-asked-for?utm_source=hashnode&utm_medium=feed) Fifteen years shipping software. The last twelve months, nothing but AI — full time, all in. What came out is seventeen tools that turn a generic coding agent into one that's actually mine. This is th 7 min read · [nott.hashnode.dev](https://nott.hashnode.dev/i-tried-to-hack-my-own-ai-agent-i-failed-that-was-the-best-result-i-could-have-asked-for?utm_source=hashnode&utm_medium=feed) [Ayesha Khan](https://hashnode.com/@ayeshakhalil) · [Jul 22](https://hashnode.com/posts/defending-llms-what-actually-works-and-where-it-still-breaks/6a613cf6032c881a9a74b390) ### [Defending LLMs: What Actually Works, and Where It Still Breaks](https://ayeshakoder.hashnode.dev/defending-llms-what-actually-works-and-where-it-still-breaks?utm_source=hashnode&utm_medium=feed) Part 2 of the LLM security series: how sandwich defenses, privilege separation & output validation actually stop prompt injection & jailbreaks. 8 min read · [ayeshakoder.hashnode.dev](https://ayeshakoder.hashnode.dev/defending-llms-what-actually-works-and-where-it-still-breaks?utm_source=hashnode&utm_medium=feed) [Ayesha Khan](https://hashnode.com/@ayeshakhalil) · [Jul 22](https://hashnode.com/posts/prompt-injection-jailbreaking-field-guide/6a602797f8d01cb875413871) ### [Breaking LLMs From the Outside In: A Field Guide to Prompt Injection and Jailbreaking](https://ayeshakoder.hashnode.dev/prompt-injection-jailbreaking-field-guide?utm_source=hashnode&utm_medium=feed) A practical taxonomy of direct/indirect prompt injection and jailbreaking techniques in LLMs. 8 min read · [ayeshakoder.hashnode.dev](https://ayeshakoder.hashnode.dev/prompt-injection-jailbreaking-field-guide?utm_source=hashnode&utm_medium=feed)