Skip to content
Request a demoGet a demo
All integrations
Claude

Claude security: Automated Prompt Masking and upload DLP.

Govern Claude web usage in real time with the Elba browser extension. Mask sensitive entities with coherent fictional values, block risky prompts or code uploads, and audit Anthropic API seats separately.

Request a demo See the workflow

Managed browser extension · Separate Anthropic API connector for seats

Capabilities

What this integration covers.

01

Mask sensitive entities inline.

Rewrite PII and confidential identifiers with coherent fictional values via EU-hosted inference (18-second fail-closed timeout).

02

Block source code and regulated uploads.

Prevent credentials, proprietary code, and sensitive text attachments (up to 5 files, 16,000 characters) from reaching claude.ai.

03

Audit Anthropic Console seats separately.

Pair browser DLP on claude.ai with Elba’s native Anthropic API connector for developer seat inventory and offboarding.

Example workflow

Protect sensitive data pasted into Claude.

How the integration supports a specific security task.

  1. 01

    Inspect prompt or upload

    The Elba browser extension inspects a prompt or text/code upload on claude.ai before send.

  2. 02

    Mask or block inline

    Matching Playbooks mask sensitive values with coherent fictional replacements or block submission.

  3. Coach the user

    The employee sees an instant explanation in the browser and security receives a structured alert.

Related capability

Explore this part of the platform.

Explore AI & browser security

Before you connect

Coverage, permissions, and requirements.

Inline Claude prompt masking and upload controls operate through the Elba browser extension on claude.ai. For Anthropic Console and enterprise seat inventory, User Access Reviews, and automated account removal, Elba also provides a separate Anthropic API connector (users and remediation).

Explore security & privacy
Does Elba support both browser Claude DLP and Anthropic API seat reviews?

Yes. The Elba browser extension governs live prompts, text/code uploads, and personal accounts on claude.ai, while Elba’s native Anthropic API connector inventories users and supports automated account removal.

How does Automated Prompt Masking protect source code or contracts in Claude?

Elba replaces detected secrets, personal identifiers, and confidential entities with coherent fictional values inline before submission, blocking the prompt if rewriting exceeds 18 seconds.

Are employee prompts stored or used for AI training?

No. Prompt analysis runs on OVHcloud AI Endpoints in Gravelines, France, stores zero raw prompt content, and never trains models on customer data.

Coverage and requirements
  • Browser coverage on claude.ai does not inventory organization seats; connect the separate Anthropic API integration for seat reviews and deprovisioning.
  • Content inspection applies to up to 5 supported text or source-code files per upload (16,000 characters each); binary files expose metadata only.
  • Automated Prompt Masking fails closed and blocks submission if rewriting exceeds the 18-second timeout.

Other applications

Review more integrations.

All integrations

ChatGPT

Automated Prompt Masking, policy blocking, and text-file DLP for ChatGPT.

Google Gemini

Distinguish Workspace vs. personal Gemini accounts and mask or block prompts.

Microsoft Copilot

Browser prompt masking for Copilot paired with SharePoint oversharing cleanup.

Discuss your requirements

Evaluate Elba for Claude.

Review the available capabilities, required permissions, and setup for your environment with our team.

Request a demo