> For the complete documentation index, see [llms.txt](https://docs.aicrisk.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.aicrisk.com/admin/ai-agents/adding-a-new-agent/advanced-settings/scanners/ai-scanners.md).

# AI Scanners

### **AI Scanners:**

<figure><img src="/files/rVBGQ4Tclwm3WmgFjiNn" alt=""><figcaption></figcaption></figure>

1. [**Toxicity Scanner**](/admin/agent-scanner-defaults/toxicity.md)
   * *Description:* The Toxicity Scanner identifies and filters out potentially harmful or offensive content from user input or generated responses.
   * *Usage:* By analyzing the text for indicators of toxicity or inappropriate language, the Toxicity Scanner helps maintain a safe and respectful conversational environment.

     <figure><img src="/files/Q76cn6XJG4BYpy7HMIMG" alt=""><figcaption></figcaption></figure>
2. [**PII (Personally Identifiable Information) Scanner**](/admin/agent-scanner-defaults/personal-information.md)
   * *Description:* The PII Scanner detects and masks personally identifiable information to protect user privacy and comply with data protection regulations.
   * *Usage:* By recognizing sensitive information such as names, addresses, or contact details, the PII Scanner ensures that such data is handled appropriately and securely.

     <figure><img src="/files/PwQm6HU0TwsrxpCaozTu" alt=""><figcaption></figcaption></figure>
3. [**Topic Scanner**](/admin/agent-scanner-defaults/topic-scanner.md)
   * *Description:* The Topic Scanner categorizes user input and facilitates topic-based response generation for improved relevance and coherence.
   * *Usage:* By analyzing the semantic context and identifying key topics within user messages, the Topic Scanner enables the agent to generate responses that align with the ongoing conversation.

     <figure><img src="/files/r51O2wsh7pdiZnzacPLo" alt=""><figcaption></figcaption></figure>
4. [**Prompt Injection Detection**](/admin/agent-scanner-defaults/prompt-injection-detection.md)
   * *Description:* Prompt Injection Detection identifies and mitigates attempts to manipulate or inject unauthorized prompts into agent interactions.
   * *Usage:* By monitoring for anomalies or unexpected changes in conversation dynamics, Prompt Injection Detection helps maintain the integrity and authenticity of interactions.

     <figure><img src="/files/VNbuGkAtnlydSywP730C" alt=""><figcaption></figcaption></figure>