> For the complete documentation index, see [llms.txt](https://docs.keeper.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.keeper.io/user-guides/developer-center/developer-center-and-ai-access.md).

# Developer Center and AI Access

End-user documentation for developer tools, integrations and AI Agent access

The Keeper Developer Center brings developer tools, integrations, and AI-agent access together in Keeper Vault. It gives developers a central place to connect Keeper with their workflows while keeping access and available capabilities under user and administrator control.

{% hint style="info" %}
Developer Center is currently in [Preview](https://docs.keeper.io/release-notes/desktop/web-vault-+-desktop-app/preview-release.md) and goes live with Vault Version 18.7
{% endhint %}

### Local tools

<figure><img src="https://914511346-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-LSGVtOTYUIkVBoYtFvK%2Fuploads%2FXCDB1mhynHMpFUCXhReM%2Fimage.png?alt=media&amp;token=d5f13cb7-cd92-471c-b11a-49e565b7b7ff" alt=""><figcaption></figcaption></figure>

The Developer Center surfaces tools that connect Keeper with day-to-day development environments. Depending on the tool and platform, these include Property Files, the Keeper MCP server, Keeper Commander, SSH Agent, KeeperDB, Secrets Manager CLI and SDKs, IDE extensions, and other developer integrations. The Center provides setup entry points and links to relevant product guidance.

### Integrations

<figure><img src="https://914511346-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-LSGVtOTYUIkVBoYtFvK%2Fuploads%2FUGn3wl5sW0Rdhl30gtqq%2Fimage.png?alt=media&amp;token=9ee19c75-8655-48ad-a1dc-afcf52f253e7" alt=""><figcaption></figcaption></figure>

The Integrations area helps developers discover ways to connect Keeper with their existing platforms and workflows, including DevOps, cloud, developer tools, and security services. Integration availability and setup requirements depend on the specific integration and the customer's Keeper services.

### AI Agent Access

<figure><img src="https://914511346-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-LSGVtOTYUIkVBoYtFvK%2Fuploads%2F2KWMV1uT9ITZ5jEzTdiy%2Fimage.png?alt=media&amp;token=6d16c67d-70d6-44a1-be37-9d7830ec8091" alt=""><figcaption></figcaption></figure>

AI Agent Access lets users connect compatible AI clients to Keeper through the Model Context Protocol (MCP). Users can enable or disable access, select which MCP tools are available, and review connected agents; administrators may control whether the feature is available to their organization. The agent's available actions are determined by the enabled tools and the user's Keeper access - not by a blanket grant to the entire Vault.

### Keeper MCP

Keeper MCP connects compatible AI clients to Keeper through the Model Context Protocol, an open standard for connecting AI applications with tools and data. With the local Keeper MCP server configured in a supported client, a user can ask an AI agent to work with Keeper data through the MCP tools that are enabled for that account.

#### User-controlled AI access

<figure><img src="https://914511346-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-LSGVtOTYUIkVBoYtFvK%2Fuploads%2FnYZAo6tXMEUcMc1BhO4U%2Fimage.png?alt=media&amp;token=09eb7c76-b96c-4419-8d1d-3042efe3f8b4" alt=""><figcaption></figcaption></figure>

AI Agent Access is configured in the Developer Center. Users can turn access on or off, choose which available tools to enable, and review or revoke connected agents. Enterprise administrators can restrict or disable AI Agent Access. The client acts within the user's available Keeper permissions and the enabled MCP tool set.

#### **Tool-specific guardrails**

<figure><img src="https://914511346-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-LSGVtOTYUIkVBoYtFvK%2Fuploads%2Fknt8rNYPvMfG0ujC1D5E%2Fimage.png?alt=media&amp;token=0964aaba-a3de-4613-a5d6-c0bbed5d52f4" alt="" width="563"><figcaption></figcaption></figure>

Access can be tailored by capability rather than granted as one broad set of permissions. Users can limit tools to selected Vault or PAM scopes, turn individual tools on or off, and see whether an action is read-only, modifies data, or is considered high-risk. For database-query access, additional settings can control whether queries may write data and the maximum number of rows returned. Some capabilities may be unavailable, disabled by an administrator, or marked as coming soon.

#### AI context in PAM settings

<figure><img src="https://914511346-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-LSGVtOTYUIkVBoYtFvK%2Fuploads%2FEdU5S4XQLXVkT0ETmOK9%2Fimage.png?alt=media&amp;token=7bb687c6-7bc0-48ee-a7d0-91de5345217f" alt=""><figcaption></figcaption></figure>

Privileged Access Management (PAM) settings can provide task-specific context to an AI workflow -for example, guidance about the purpose of a managed system or how an approved task should be approached. This context helps an agent interpret a request without treating free-text instructions as authorization. Access to PAM resources remains governed by Keeper permissions, administrator policy, and the enabled MCP capabilities.


---

# 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://docs.keeper.io/user-guides/developer-center/developer-center-and-ai-access.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.
