https://sacatani.com/ Simply curious about the world, constantly building and breaking things for fun. en https://sacatani.com/writing/ai-pentesting-continuous-testing/ https://sacatani.com/writing/ai-pentesting-continuous-testing/ Wed, 23 Sep 2026 12:00:00 GMT I used to spend most of my time testing web applications. My first look at AI pentesting products suggests the market is moving from point-in-time reports toward continuous, context-rich security testing. https://sacatani.com/writing/mcp-tools-trust-too-much/ https://sacatani.com/writing/mcp-tools-trust-too-much/ Tue, 22 Sep 2026 12:00:00 GMT A first-person walkthrough of two MCP attack labs: a Xata-style read-only SQL bypass and a GitHub public-issue injection that exposed the difference between model safety and application security. https://sacatani.com/writing/mcp-confused-deputy-problem/ https://sacatani.com/writing/mcp-confused-deputy-problem/ Sun, 20 Sep 2026 12:00:00 GMT A hands-on MCP lab showed how a valid tenant key can still retrieve another tenant's project when the server never checks resource ownership. https://sacatani.com/writing/mcp-filesystem-prefix-bypass-response-poisoning/ https://sacatani.com/writing/mcp-filesystem-prefix-bypass-response-poisoning/ Sun, 20 Sep 2026 12:00:00 GMT Two MCP labs showed different failures at the same boundary: a filesystem server trusted a string prefix, while a facts server poisoned the agent's next tool call. https://sacatani.com/writing/mcp-security-new-attack-surface/ https://sacatani.com/writing/mcp-security-new-attack-surface/ Sat, 19 Sep 2026 12:00:00 GMT MCP makes it easier for AI systems to use external tools. It also connects model behaviour to permissions, private data, and real actions. https://sacatani.com/writing/poisoning-mindfulchat-memory/ https://sacatani.com/writing/poisoning-mindfulchat-memory/ Thu, 17 Sep 2026 12:00:00 GMT Lakera's MindfulChat challenge made persistent memory feel like a real application attack surface, not just a model prompt. https://sacatani.com/writing/interactive-playground-llm-security/ https://sacatani.com/writing/interactive-playground-llm-security/ Wed, 16 Sep 2026 12:00:00 GMT I worked through an interactive OWASP LLM Top 10 playground and found a useful starting map, a few strong concepts, and several labs that needed more explanation. https://sacatani.com/writing/web-llm-attacks-deeper/ https://sacatani.com/writing/web-llm-attacks-deeper/ Tue, 15 Sep 2026 12:00:00 GMT Indirect prompt injection and insecure output handling showed me that an LLM can turn ordinary application content into an attack path. https://sacatani.com/writing/web-llm-attacks/ https://sacatani.com/writing/web-llm-attacks/ Tue, 15 Sep 2026 12:00:00 GMT Mapping an LLM's tools first turned two PortSwigger labs into the same repeatable web-testing method. https://sacatani.com/writing/secure-systems-for-mobile-applications/ https://sacatani.com/writing/secure-systems-for-mobile-applications/ Tue, 15 Sep 2026 12:00:00 GMT I spent a year at EY breaking mobile apps, then built one at MacroScope and had to defend against everything I used to find. The one lesson that carried across both sides. https://sacatani.com/writing/senior-and-a-red-team-from-scratch/ https://sacatani.com/writing/senior-and-a-red-team-from-scratch/ Fri, 11 Sep 2026 12:00:00 GMT How I got onto a red team that did not exist yet, and the printer that got us onto a network that thought it was locked down. https://sacatani.com/writing/joining-ey-as-an-associate-pentester/ https://sacatani.com/writing/joining-ey-as-an-associate-pentester/ Fri, 11 Sep 2026 12:00:00 GMT I applied for one job, had one interview, and got it. What the first year of pen testing actually looked like.