Developers are moving faster than ever with AI. Cursor is redefining how software gets built, and 1Password is redefining how teams secure access to SaaS and AI. Today, we are announcing a new integration that brings these two worlds together in a way that keeps development speed high and credential risk near zero.

1Password has partnered with Cursor to build a Hooks Script that gives developers a secure, just-in-time way to ensure required secrets are made available to Cursor’s AI agents via 1Password Environments. The result is an AI-native development workflow where secrets are never hardcoded, raw credentials are never handled directly by AI agents, and secure access becomes a natural part of writing and running code.

This functionality is available today as a first step and lays the foundation for a broader set of secure developer workflows we intend to build together.

Why this matters

Developers should never have to paste tokens into config files or store long-lived credentials on disk. AI agents in their editors should not have unrestricted access to secrets either.

By integrating 1Password with Cursor Hooks, we are making 1Password the secure source of secrets for Cursor. When the Cursor agent needs to run a command, call an API, or perform an action that requires a credential, the required secret can be made available at runtime through 1Password, only when authorized by the user. 

No plaintext keys committed to disk or source code. No hard-coded environment variables. No tokens lingering in history. Everything is made available securely via 1Password and governed by the access policies your team already relies on. Furthermore, the project owner can configure 1Password secrets management, helping ensure secure practices are consistently followed across the team.

This provides teams with a clear path to adopt AI-powered development while maintaining a strong security posture.

About Cursor

Cursor is an AI-powered IDE built on Visual Studio Code that adds deeply integrated AI assistance throughout the development workflow. Developers can write or modify code using natural language, search across large projects by meaning, and perform structured, multi-line edits with a simple prompt.

Cursor also provides a powerful integration layer through the Model Context Protocol (MCP). This enables the editor to interact directly with APIs, databases, and external tools within the development environment. Cursor Hooks extend this further by enabling teams to run scripts automatically at specific points in an AI-assisted workflow.

This new Hooks system is the cornerstone of our integration.

What Cursor Hooks enable

Cursor Hooks allow teams to configure a file called hooks.json at the project, user, or system level. This file outlines what should occur at specific lifecycle stages of an AI-assisted interaction. For example, before Cursor runs code, executes a command, or interacts with a tool, Cursor invokes the Hook Script to prepare the right environment.

Our new Hooks Script makes 1Password the secure source of truth for secrets, configurations, and credentials that Cursor might need.

Here is how it works at a high level:

1. Before the Cursor agent runs any shell commands, the Hook Script is invoked.

2. The script verifies that all required locally mounted .env files from 1Password Environments are properly configured, ensuring commands that depend on them run without issue.

3. After the script checks your .env files, it either runs the command or returns an error message to help you fix your 1Password Environments setup.

4. When a process requests access, 1Password prompts the user to authorize and makes the secret available in memory for the runtime session. It never touches disk or Git history.

This creates a secure, repeatable workflow where developers do not need to manually copy credentials, rotate tokens, or worry about accidental exposure.

Explore the full 1Password Environments documentation for Cursor Hooks.

What is available today

With 1Password, Cursor users can:

• Use 1Password as the secure credential store for AI-driven tasks in Cursor.

• Configure Cursor Hooks that validate required .env files managed by 1Password at runtime, ensuring secrets are available only when needed and governed by 1Password.

• Version control Hooks configuration files without exposing any sensitive values.

• Enable AI-powered development in Cursor without changing existing 1Password policies, vaults, or user permissions.

This initial functionality is intentionally simple: it keeps secrets out of code and provides developers with a safer way to allow Cursor to perform tasks that require credentials.

What we are building next

The work launching today is the foundation for a deeper collaboration. In the coming months, we plan to expand the integration to support:

• Richer policies and permissions that allow teams to define granular, task-specific access rules for AI agents.

• Broader support for MCP integrations so that Cursor can interact with external APIs and services entirely through 1Password-mediated access.

• Automated secret rotation for AI-driven workflows.

• Enhanced audit visibility to enable security teams to monitor how AI agents access credentials throughout the development lifecycle.

Our goal is to create the first AI native development environment where secure access is not an afterthought but a built-in part of the workflow.

Accelerate securely with 1Password and Cursor

AI is transforming how software gets built, but speed only helps when teams can trust the workflows behind it. By integrating 1Password with Cursor Hooks, we are eliminating one of the biggest sources of risk in modern development: uncontrolled secrets.

Developers get a faster workflow. Security teams get centralized control. And AI agents get only the access they need, exactly when they need it.

This is just the beginning. We are excited to continue building with the Cursor team and help shape the future of secure AI-assisted development. You can get started with the integration here.