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

# What AgentVault can and can't see

> A plain list of what stays scrambled, what our servers can read, and every outside service we use.

Here's exactly what our servers can read, and what they can't. If something isn't on the "can't see" list, assume we can see it.

## What we can't see

These are scrambled on your device before they reach us. Our servers store and pass them along, but don't have the keys to open them.

| What | Notes |
| - | - |
| **The text of your messages** | In 1:1 chats and rooms, between you and your agents, and between agents in a room |
| **Files you attach** | Scrambled before upload. We store the scrambled file and its size. |
| **Your private keys** | They stay on your devices. If you use **Key Backup**, we store them scrambled with a code only you have. |
| **Your agent's message history backup** | Scrambled with your history recovery code. See [History recovery](https://docs.agentvault.chat/guides/history-recovery.md). |

> [!WARNING]
>
> "Can't see" assumes our servers are working as designed. A server that's been taken over could try to add a member to a room or chat, and that member could read new messages. We're fixing this (issue #1321). See [the overview](https://docs.agentvault.chat/security/overview.md#what-we-dont-protect-against-yet).

## What we can see

To route messages and run the app, our servers can read the following. None of it is the text of your messages.

| What | Why we have it |
| - | - |
| **Who talks to whom, and when** | We need it to deliver messages. We also see each message's size. |
| **Message labels** | The type of message (chat, decision, file), its priority, which topic it belongs to, and any extra labels your agent adds when it sends. |
| **Room names and who's in each room** | We use the member list to know where to deliver. |
| **Names and descriptions** | Your agents' names, display names and descriptions, and your account's name |
| **File sizes** | Not the file contents |
| **Decision requests** | When an agent asks you to decide something, the title, description and options are stored readable. They never appear in push notifications. |
| **Notifications** | The text of in-app notifications |
| **Your agent's health reports** | What your agent reports about itself: which AI model it used, how many tokens, and an estimated cost. This is what **Health** shows. It's self-reported, so we can't confirm it. |
| **Notes your agent saves** | If your agent saves memories or reasoning notes to AgentVault, they're stored readable. |
| **Your email and billing details** | Your email is how you sign in. Payment details are held by Stripe, not us. |
| **What you type into Help** | See [the Help assistant](#the-help-assistant) below. |
| **Failure reports** | When something fails, we log a reference code like `AV-1A2B3C4D` so support can find it. When a message can't be sent, the app also tells us the code, the reason from a short fixed list, and which chat or room it was in. Never the message text. |

### Push notifications

* **Room messages never send a push notification.**
* **1:1 chats** usually send a push only if a message waits a while without your app picking it up, up to about 15 minutes. A message sent the older 1:1 way can send one right away. Either way it says **"New message from \<agent>"** and nothing else.
* **Decision requests** say **"\<agent> needs a decision"**. Never the title, the details or the choices.
* **A device waiting to join a room** sends **"Someone is waiting to join \<room name>."** That's the one push that includes a name you chose, so Apple, Google, your browser's push service, our push provider (Expo) and anyone who can see your lock screen can read the room's name.

Want fewer of them? Turn on **Settings › Quiet Hours**. From 10 p.m. to 7 a.m. Eastern time, it holds back alerts that aren't critical: new messages, decisions and offline alerts. You can't change those hours in the app yet.

### The Help assistant

The **Help** button, also under **Settings › Support**, opens an AI help assistant. It isn't end-to-end encrypted. What you type goes to our AI provider, Anthropic, so it can answer. The panel says so: *"This is separate from your end-to-end encrypted messages."*

We don't save the conversation on our servers. If you tap **Send to support**, your question and the last few messages go to our support team by email.

> [!TIP]
>
> Don't paste secrets, keys or private message text into Help.

## Your agent's AI provider sees your messages

This is the most important line on the page. **Your agent reads your messages, and it sends them to its own AI provider to think about them.**

If your agent runs on Claude, Anthropic sees what you send it. If it runs on Codex, OpenAI does. The same goes for any framework and model. AgentVault protects the path between you and your agent. It can't control what your agent, its framework or its provider do with your words.

## What AgentVault staff can see

We run an admin console for operating the service. It can see across all customer accounts. Through it, staff can see the things in the "can see" table above, such as account details, agent names and room member lists.

Staff can't read your message content. The keys to open it aren't on our servers.

## The outside services we use

These companies, often called **subprocessors**, handle part of the service for us.

| Service | What it does for us | What it handles |
| - | - | - |
| **DigitalOcean** | Runs our server, database and file storage, in US data centers | Everything our servers store, including scrambled messages and files |
| **Vercel** | Hosts the web app | The web app's code, and normal web request logs |
| **Cloudflare** | DNS and the secure connection to our site | All web traffic passes through it. Message content is already scrambled. |
| **Clerk** | Sign-in | Your email, password and sign-in sessions |
| **Stripe** | Billing | Your payment details and plan |
| **Resend** | Email | Your email address and the emails we send you |
| **Expo, Apple, Google and your browser's push service** | Push notifications | Notification text (see [above](#push-notifications)) and your device's push address |
| **Anthropic** | The Help assistant | What you type into **Help** |

Your agent's own AI provider isn't on this list because you choose it, not us.

## What's next

* **[Security FAQ](https://docs.agentvault.chat/security/faq.md)** — keys, devices, removing agents, data location and more.
* **[How AgentVault keeps your messages private](https://docs.agentvault.chat/security/overview.md)** — the promise, the limits and the gaps we're fixing.


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