> For the complete documentation index, see [llms.txt](https://developer.harness.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://developer.harness.io/ai-security/3.0/ai-security.md).

# AI Security

AI Security provides you with visibility into AI APIs and MCP assets across environments and understand how they interact with your applications. You can monitor threats targeting AI endpoints, investigate related issues, and test AI endpoints for vulnerabilities such as prompt injection and sensitive data exposure. These capabilities help you manage AI risk using the same security workflows applied to your APIs.

<figure><img src="/files/aOUkeDraML45SsrcSEdH" alt=""><figcaption></figcaption></figure>

### Get Started

<table data-view="cards"><thead><tr><th></th><th></th><th data-hidden data-card-target data-type="content-ref"></th></tr></thead><tbody><tr><td><strong>AI Security Overview</strong></td><td>Understand how AI Security helps you discover AI assets, monitor threats, and test AI endpoints in your application.</td><td><a href="/spaces/WkhxazUhF4HZSnOgyi26/pages/PuuCPInELqR8Q6RdAO08">/spaces/WkhxazUhF4HZSnOgyi26/pages/PuuCPInELqR8Q6RdAO08</a></td></tr><tr><td><strong>AI Security Agent Support Matrix</strong></td><td>Understand which agents support AI Security and how to enable capabilities, such as AI asset discovery and threat activity.</td><td><a href="/spaces/WkhxazUhF4HZSnOgyi26/pages/rgr4GRCYpcQUX7F2x7E5">/spaces/WkhxazUhF4HZSnOgyi26/pages/rgr4GRCYpcQUX7F2x7E5</a></td></tr><tr><td><strong>AI Security Dashboard</strong></td><td>View AI assets, identified issues, threat activity, and overall AI security posture from a centralized dashboard.</td><td><a href="/spaces/WkhxazUhF4HZSnOgyi26/pages/FHLIxAmgCtEa6Ie5X8IB">/spaces/WkhxazUhF4HZSnOgyi26/pages/FHLIxAmgCtEa6Ie5X8IB</a></td></tr></tbody></table>

### AI Discovery

<table data-view="cards"><thead><tr><th></th><th></th><th data-hidden data-card-target data-type="content-ref"></th></tr></thead><tbody><tr><td><strong>AI Assets</strong></td><td>Discover AI inventory consisting of AI APIs and MCP assets, such as tools, servers, resources, and prompts in your application.</td><td><a href="/spaces/WkhxazUhF4HZSnOgyi26/pages/zxq2477JFxLvbhSyI8KI">/spaces/WkhxazUhF4HZSnOgyi26/pages/zxq2477JFxLvbhSyI8KI</a></td></tr><tr><td><strong>AI Asset Details</strong></td><td>Understand how AI asset details help you monitor activity, analyze performance, and identify risks in your AI assets.</td><td><a href="/spaces/WkhxazUhF4HZSnOgyi26/pages/QSR4INUUyx5LoiXFLHlU">/spaces/WkhxazUhF4HZSnOgyi26/pages/QSR4INUUyx5LoiXFLHlU</a></td></tr><tr><td><strong>MCP Risk Score</strong></td><td>Understand risk scores assigned to AI assets based on exposure, behavior, and detected issues.</td><td><a href="/spaces/WkhxazUhF4HZSnOgyi26/pages/s445fgjdrkkKHJpnflSi">/spaces/WkhxazUhF4HZSnOgyi26/pages/s445fgjdrkkKHJpnflSi</a></td></tr></tbody></table>

### AI Testing

<table data-view="cards"><thead><tr><th></th><th></th><th data-hidden data-card-target data-type="content-ref"></th></tr></thead><tbody><tr><td><strong>(Beta) AI Security Testing</strong></td><td>Test AI APIs and applications for OWASP LLM Top 10 issues, such as prompt injection and sensitive data exposure.</td><td><a href="/spaces/WkhxazUhF4HZSnOgyi26/pages/3DnXogrWKkQGI1ZaVU61">/spaces/WkhxazUhF4HZSnOgyi26/pages/3DnXogrWKkQGI1ZaVU61</a></td></tr></tbody></table>