How to connect Lakera Guard to Claude Cowork

Cowork is Anthropic's AI agent for knowledge work. Think of it as Claude Code for everything else. It works autonomously with your computer, local files, and applications to accomplish complex tasks.

This guide walks you through the easiest and most secure way to connect your Lakera Guard account to Cowork via Composio Connect, enabling it to screen prompts for injection attempts, list active Lakera Guard projects, flag unsafe chatbot responses before sending, and more such actions on your behalf without compromising your account security.

Connecting Lakera Guard to Claude Cowork

1. Open Customize

In Claude Desktop, click Customize in the left sidebar, then select Connectors and click the + icon at the top.

2. Add the Composio MCP server

Click Add custom connector and paste in the Composio MCP server URL:

Claude Cowork is Anthropic's agent for general knowledge work. It can use your computer, files, and connected applications to complete longer-running tasks across your work tools.

With Composio Connect, Cowork can securely access apps like Lakera Guard through MCP without you sharing account credentials directly with the agent.

The Lakera Guard MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Lakera Guard account. It provides structured and secure access to content screening, detector evaluations, guardrail policies, and projects, so your agent can screen interactions, assess risks, configure policies, manage projects, and review safety settings on your behalf.

• AI interaction screening: Have your agent screen the latest LLM or agent interaction and return the policy decision before your application continues.
• Focused risk evaluation: Direct your agent to evaluate content against configured detectors and review the confidence result from each detector.
• Guardrail policy management: Let the agent create, review, update, or delete policies that combine built-in detectors with custom guardrails.
• Safety project management: Instruct your agent to create, inspect, update, or delete projects while assigning policies, metadata, and detection or flagging behavior.
• Configuration review: Have your agent retrieve policy and project details to check detector settings, custom guardrails, assigned projects, metadata, and configured actions.