Skip to content

Cursor integration

Stop secrets and customer data leaking through Cursor

Maskflare hooks for Cursor stop prompts that contain API keys or customer data and mask tool output before it reaches the model, using the same rules as the rest of your company.
What Maskflare covers
  • Prompts: blocked when they contain a secret or a blocked data type
  • Tool output that Cursor's hook API exposes
  • Your organization's rules, including 232 credential formats

Cursor's agent mode reads files and runs commands to fix a bug or write a feature. That's the point of it, and it's also how a database URL or a customer export ends up in a prompt.

maskflare-hook adds Maskflare's checks to Cursor's hooks, in the IDE and the CLI: prompts with secrets are stopped, shell and tool output is masked, and sensitive files are protected. The desktop agent installs and maintains the hook for every developer.

Set up Cursor with Maskflare

Maskflare installs into Cursor's hook system to stop prompts that carry secrets and mask what its tools return, within what Cursor's hook API allows.

Install the Cursor hook
pip install maskflare
maskflare-hook install cursor

What gets masked

Good to know

Frequently asked questions

Is Cursor coverage the same as Claude Code?

No. Cursor's hook API exposes less, so the Cursor hook covers prompts and the tool events Cursor makes available.

Can we roll it out to the whole team?

Yes. Install it in your developer image or with your usual setup scripts, and manage the rules in the Maskflare console.

Does it also cover Codex CLI?

Yes. Run maskflare-hook install codex.

Other integrations

See it on your own data

Try Maskflare free.
Bring your hardest prompts.

28-day free trial or 10,000 requests, whichever comes first. Point one app at the gateway or enroll a few laptops,
and watch what gets masked. Or talk to us first.