> For the complete documentation index, see [llms.txt](https://help.nightfall.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.nightfall.ai/data-exfiltration-prevention/ai-agent-security/ai-governance/auditability-and-control/setup-and-installation.md).

# Setup & Installation

AI Agent Security has two installs. They are not substitutes.

* **Claude Inference Hooks** cover Claude Enterprise prompts before the model runs. No endpoint agent and no MDM. [Connect Inference Hooks](/data-exfiltration-prevention/ai-agent-security/ai-governance/auditability-and-control/setup-and-installation/connect-inference-hooks.md).
* **Device hooks** cover Claude Code, Cursor, and VS Code on the laptop, and can stop a shell command or a tool call before it runs. Those need the agent and a separate hooks deployment.

### Before device hooks

1. **Nightfall endpoint agent v1.2.12.11+** installed and running. The hooks call `nightfall-hook-relay` by name. Without the agent, the hooks resolve to nothing. The agent installer does not deploy hook settings.
2. **Nightfall agent configuration profile v3+**, deployed with the agent through your MDM (Jamf, Mosyle, Kandji, and the other agent-deployment guides). This profile grants the agent its OS permissions. If it is missing, the agent shows a missing-permissions error and you will not see local or remote MCP servers.
3. **An MDM scope**, the device group, Blueprint, Smart Group, or assignment that will receive the hooks deployment.
4. **A current client** on managed devices (Claude Code, Cursor, or VS Code, depending on which hooks you deploy).

The agent profile is not the Claude Code file `nightfall-hooks.mobileconfig`. Most Claude Code fleets never deploy that file. Use it only when a `com.anthropic.claudecode` managed-preferences profile is already the winning settings source. See Nightfall Hooks for Claude Code.

Inference Hooks do not use this checklist.

### Deploy device hooks

Pick the guide for each client. The `README.md` in each deployment zip matches its guide.

* Nightfall Hooks for Claude Code. macOS uses `payloads/nightfall-hooks.json`. Windows uses `payloads/nightfall-hooks.windows.json`, staged under the name `nightfall-hooks.json`. Check for a higher-priority Claude Code managed source before using the file drop.
* Nightfall Hooks for Cursor.
* [Nightfall Hooks for VS Code Copilot](/data-exfiltration-prevention/ai-agent-security/ai-governance/auditability-and-control/setup-and-installation/nightfall-hooks-for-vs-code-copilot.md). The install script copies the hook file into each user's `~/.copilot/hooks` folder. On Windows, stage `payloads/nightfall.windows.json` as `nightfall.json`.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://help.nightfall.ai/data-exfiltration-prevention/ai-agent-security/ai-governance/auditability-and-control/setup-and-installation.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.