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

# Guides

> Pick the guide that matches what you are building: enforce content, bring your own policy, audit a backlog, or run it in production.

Each guide solves one job and ends with code you can run. Find your goal below. For a shorter list by product surface, see [Use Cases](https://docs.zerodrift.com/use-cases.md).

New here? Send a request with the [Quickstart](https://docs.zerodrift.com/quickstart.md), then read [Concepts](https://docs.zerodrift.com/concepts.md) for the words these guides use and [Choose a model](https://docs.zerodrift.com/models.md) to pick Mini, Flagship, or Max.

## Enforce content

Check content before it reaches a customer, a regulator, or another system.

| I want to | Guide | Model in the guide |
| - | - | - |
| Decide what my code does on Pass, Fix, Block, and Escalate | [Act on Each Verdict](https://docs.zerodrift.com/guides/act-on-verdicts.md) | Flagship |
| Check every reply my chatbot generates before the customer sees it | [Enforce a Chatbot](https://docs.zerodrift.com/guides/enforce-a-chatbot.md) | Flagship |
| Check what my AI agents send, including messages between agents | [Agent-to-Agent Traffic](https://docs.zerodrift.com/guides/agent-to-agent.md) | Flagship |
| Check documents, articles, and other long messages | [Long-Form Content](https://docs.zerodrift.com/guides/long-form-content.md) | Flagship |

## Use your own policy

Enforce your rules, not only the managed rulepacks.

| I want to | Guide | Model in the guide |
| - | - | - |
| Turn my policy document into enforceable rules | [Import a Policy](https://docs.zerodrift.com/guides/import-a-policy.md) | Max |
| Train an adapter on my policy and enforce against it | [Train an Adapter](https://docs.zerodrift.com/guides/train-an-adapter.md) | Flagship |

## Audit content

Check a backlog or batch after the fact, or hand the work to a coding agent.

| I want to | Guide | Model in the guide |
| - | - | - |
| Hand a file of content to my coding agent and get a verdict for every item | [Validate at Scale with AI](https://docs.zerodrift.com/guides/validate-at-scale-with-ai.md) | Flagship |

## Run it in production

Make the integration safe to depend on, and keep a record of what it did.

| I want to | Guide | Model in the guide |
| - | - | - |
| Retry the right failures, respect rate limits, and choose fail-open or fail-closed | [Handle Errors and Retries](https://docs.zerodrift.com/guides/handle-errors.md) | Flagship |
| Pull the verdict log for review, reporting, and my own retention | [Export Activity](https://docs.zerodrift.com/guides/export-activity.md) | Not applicable |

## Tools and setup

| I want to | Page |
| - | - |
| Generate a typed Python or TypeScript client | [SDKs](https://docs.zerodrift.com/generate-sdk.md) |
| Pull a policy from Notion, Linear, Confluence, or Google Drive | [MCP Connectors](https://docs.zerodrift.com/connecting-mcp.md) |
| Let a coding agent read these docs | [For AI Agents](https://docs.zerodrift.com/for-agents.md) |
| See every endpoint and field | [API reference](https://docs.zerodrift.com/api-reference/introduction.md) |


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.