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

# Permissions

> How agent access is scoped and credentials stay secure

Giving an agent real access to your tools is a big step. This page explains the security model behind every connection.

## The transparent proxy

All of an agent's outbound traffic flows through a **proxy** between the sandbox and the internet. Credentials are applied there. When the agent makes a request to a connected service, the proxy attaches the right token **on the wire**. The agent's own code and prompts never contain it.

```
Agent sandbox    Proxy    External service
                     
            attaches your credential here,
            never inside the agent
```

The result: the agent has the **capability** to act in a service without ever holding the **credential**.

## Why it's designed this way

Large language models shouldn't be trusted with raw secrets. A token in a prompt could surface in a log, a message, or a memory file. Skydive keeps credentials out of the agent entirely and injects them at the network layer. That removes the whole class of risk.

<CardGroup cols={2}>
  <Card title="Secrets never leak" icon="lock">
    Tokens aren't in prompts, logs, or memory. They live outside the agent's reach.
  </Card>

  <Card title="Scoped per service" icon="filter">
    Each connection is limited to the service and permissions you granted.
  </Card>

  <Card title="You control the grant" icon="sliders">
    Authorize narrowly or broadly, and revoke at any time.
  </Card>

  <Card title="Isolated sandboxes" icon="box">
    Each agent runs in its own machine. One agent's access doesn't bleed into another's.
  </Card>
</CardGroup>

## Scoping access

You decide how much an agent can reach:

* **Per service**: connect only the tools a given agent needs.
* **Per resource**: for services like [GitHub](/docs/integrations/github), grant specific repositories rather than everything.
* **Per permission**: connections request only the scopes required for the work.

## Revoking access

Remove a connection or rotate a key at any time. The agent loses that access on its next request. No lingering tokens, no manual cleanup.

## A note on memory

[Memory](/docs/agents/memory) lives in the agent's [git repository](/docs/agents/repo), scoped to your workspace. It's shared knowledge for everyone with access to the agent. Treat it like a team wiki, not a vault. Keep genuine secrets in [secrets](/docs/integrations/secrets), not in things you tell the agent to remember.

<Card title="Connect a service securely" icon="plug" href="/docs/integrations/oauth">
  See the authorization flow in practice.
</Card>