Guides

PII redaction strips personal data before an AI model can read it. How redaction, masking, and tokenization differ, and how SealGate masks PII in tool results.

AI guardrails limit what an AI system can read, say, or do. How LLM and agent guardrails work, and why deterministic tool-call rules survive prompt injection.

Claude computer use lets Claude control a desktop like a person, by screenshot, click and type. How it compares with MCP tools, and when each is the better fit.

AI data loss prevention (AI DLP) controls what AI agents can read and where that data can go. How it differs from traditional DLP, and how SealGate enforces it.

Lethal trifecta - the three-capability threat model behind AI-driven data exfiltration - private data access, untrusted content exposure, and external communication - and how SealGate blocks it.

An MCP proxy server sits between an MCP client and one or more upstream MCP servers, merging their tools into one endpoint and routing every call to the right server. What it is, how to run local (stdio) and remote (HTTP) proxies, and why you put one in front of your MCP servers.

AI agent governance is the set of controls that decide which AI agents can act, on whose behalf, on which tools and data, and with what record left behind.

Prompt injection is the top security threat to AI agents - what it is, how direct and indirect prompt injection attacks work, and how SealGate contains them by blocking the Lethal Trifecta rather than trying to detect the injection.

An MCP gateway is one governed entry point between AI agents and MCP servers. How it works, how to choose one, and the best MCP gateways compared side by side.