About Content Security

Content Security is an Access PolicyAccess Policy: Access Policies define, enforce, and audit access between Client and Server Workloads by cryptographically verifying workload identity and contextual factors rather than relying on static secrets.Learn more component that inspects content and enforces the verdict (allow, block, or transform) an inspection service returns. When a policy includes Content Security, Aembit sends eligible content to the inspection service and applies the returned verdict before the request continues.

Content Security inspects Model Context Protocol (MCP)Model Context Protocol: A standard protocol for AI agent and server interactions that defines how AI assistants communicate with external tools and data sources.Learn more(opens in new tab) traffic that flows through the Aembit MCP Identity GatewayMCP Identity Gateway: A component that brokers MCP traffic between MCP clients and target MCP servers, validating authorization and presenting Aembit-managed credentials on each request.Learn more. In the Access Policy Builder, Content Security is an optional component positioned between Access Conditions and Credential Providers. It adds inspection to the request path without changing how the policy’s other components authorize access.

Start configuring Content Security See Content Security in the User Guide

→

How Content Security works

Content Security inspects MCP traffic at three checkpoints on the request path:

• Tool listings - The set of tools an MCP server advertises to a client.
• Tool call inputs - The arguments a client sends when it calls a tool.
• Tool call outputs - The results a tool returns to the client.

At each checkpoint, the inspection service evaluates the content and returns one of three verdicts:

• Allow - The content passes inspection, and Aembit forwards it unchanged.
• Block - The content violates a policy, and Aembit blocks it.
• Transform - The inspection service returns modified content, such as content with sensitive data redacted, and Aembit forwards the transformed content in place of the original.

You configure the inspection service on the Content Security component. The available service is CrowdStrike AI Detection and Response (AIDR)AIDR: CrowdStrike AI Detection and Response (AIDR) is a content inspection service that scans MCP tool listings, tool call inputs, and tool call outputs in real time and returns an allow, block, or transform verdict to Aembit.Learn more. It scans MCP content in real time and returns the verdict that Aembit enforces. See Add CrowdStrike AIDR to a policy to configure the integration.

The following diagram shows how the Aembit MCP Identity Gateway applies Content Security inspection to MCP traffic:

Inspection scope

Content Security evaluates one MCP message at a time. Each checkpoint produces an independent verdict, and that verdict governs only the message it applies to.

Inspection is per MCP message

Content Security evaluates each MCP message on its own. A verdict covers only the single tool listing, tool call input, or tool call output the inspection service evaluated, not the conversation, session, or task that message belongs to. Blocking one message doesn’t block the messages that follow it. If an AI agent retries a blocked call with different arguments and the new message passes inspection, Aembit forwards it.

Content Security guarantees that no inspected message reaches its destination without a verdict. It doesn’t correlate messages or track a task across the requests an AI agent makes to complete it. An agent that can’t finish a task one way commonly tries another. When an alternate attempt produces content that passes inspection, Aembit allows that content.

CrowdStrike AIDR and the CrowdStrike Access Condition

Aembit integrates with two different CrowdStrike products through two different Access Policy components. CrowdStrike AIDR is a Content Security inspection service that evaluates MCP content. The CrowdStrike Access Condition evaluates endpoint posture, such as the reported state of a Client Workload’s endpoint. Configuring one doesn’t enable the other.

Data sharing and privacy

When a policy includes Content Security, Aembit sends the inspected MCP content to the inspection service. This content includes the MCP tool listings, tool call inputs, and tool call outputs on the request path. Aembit also sends request metadata that identifies the request, such as the user and agent identifiers and the application and server names involved. Review the inspection service’s data handling before you add Content Security to policies that carry sensitive content.

Benefits of using Content Security

• Content-level control - Inspects the content of MCP tool listings, tool call inputs, and tool call outputs, beyond the identity and context that other Access Policy components evaluate.

• Automated enforcement - Applies the inspection service’s allow, block, or transform verdict on each matching request without manual review.

• Content transformation - Transform verdicts return modified content, such as content with sensitive data redacted, in place of the original.

• Policy-scoped inspection - Applies only to the policies you add it to, so you can target inspection where it’s needed.