> ## Documentation Index
> Fetch the complete documentation index at: https://anthale.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Anthale Guardrails for Runtime Risk Control

> Review Anthale guardrail families and move from control selection into prompt injection, data leakage, moderation, topic control, and link guidance.

Guardrails are the risk-specific controls Anthale runs against content and behavior patterns. Policies decide which guardrails apply and how their results are enforced.

## What guardrails do

* Detect prompt injection and instruction takeover attempts.
* Catch sensitive data exposure.
* Moderate harmful or disallowed content.
* Enforce scope boundaries with topic and link controls.
* Check trust and scope at runtime.

## How guardrails and policies relate

* Guardrails are the evaluators and rule sets.
* Policies decide when those guardrails run and what happens after a hit.
* Your application still owns final runtime enforcement in code.

Each guardrail overview page links to focused deep-dive material so this page can stay explanatory and the detailed pages can stay specific.

## Available guardrails

<Columns cols={2}>
  <Card title="Prompt Injection Protection" icon="syringe" href="/docs/learn/guardrails/prompt-injection">
    Detect attempts to seize instruction control or override system behavior.
  </Card>

  <Card title="Data Leakage Prevention" icon="database" href="/docs/learn/guardrails/data-leakage">
    Catch sensitive content before it crosses the wrong runtime boundary.
  </Card>

  <Card title="Content Moderation" icon="file-text" href="/docs/learn/guardrails/content-moderation">
    Moderate harmful or disallowed content categories on input and output.
  </Card>

  <Card title="Topic Control" icon="eye" href="/docs/learn/guardrails/topic-control">
    Keep an assistant inside the subjects it is allowed to handle.
  </Card>

  <Card title="Link Control" icon="link" href="/docs/learn/guardrails/link-control">
    Decide which domains are allowed, blocked, or treated as malicious.
  </Card>
</Columns>

## Common combinations

* Customer-facing chat: prompt injection + data leakage + content moderation.
* RAG systems: prompt injection + data leakage + link control.
* Brand-sensitive assistants: topic control + link control + content moderation.

## How to choose

* Start with the risks that create the biggest customer or security impact.
* Add policy controls where the boundary is explicit.
* Add monitoring first in lower-risk paths, then move toward blocking or redaction.

## What to read next

Start with the guardrail that matches the risk you are trying to reduce. If the problem is control takeover, go to [Prompt Injection Protection](/docs/learn/guardrails/prompt-injection). If the problem is sensitive data exposure, go to [Data Leakage Prevention](/docs/learn/guardrails/data-leakage). If the problem is product scope, go to [Topic Control](/docs/learn/guardrails/topic-control).