Aceby Black Label

AI Safety Compliance Detection · Security & Identity Tools

Lakera Guard
+ Ace

Lakera Guard screens AI application interactions for security and safety risks and provides Enterprise APIs for managing guardrail policies and projects.

Connect it to your Ace account. Use it in Ace and Agent Board, or through your own code. 10 actions; anything that acts waits for your yes.

Open Agent Board

What Ace can do in Lakera Guard

2 actions that read, 8 that act.

Three ways to play it

Ask Ace

“Create Policy” on your iPhone, Mac or Windows. Ace finds the action, reads what it needs, and asks before it does anything.

Get Ace

Any AI chat

Add Ace's MCP link to Claude, ChatGPT, Cursor or VS Code and Lakera Guard is there, with every other app you connected.

Set up MCP

Your code

One API key. ace.tools.run("LAKERA_GUARD_CREATE_POLICY", {…}) in JavaScript or Python.

Get the SDK

Also in the deck

Questions

How do I connect Lakera Guard to Ace?

Open tryacee.com/integrations/lakera_guard while signed in to Ace and choose Connect Lakera Guard. You sign in on Lakera Guard's secure connect page; Ace never sees your password. The connection then works in Ace on iPhone, Mac and Windows, in AI chats over MCP, and through the Ace API.

Will Ace send or change things in Lakera Guard without asking?

No. When you ask Ace's assistant to do something, actions that only read run right away. Anything that sends, posts, creates, changes or deletes is described to you first and runs only after you reply yes.

Can I use Lakera Guard from Claude or ChatGPT?

Yes. Add Ace's MCP server (tryacee.com/mcp) to Claude, ChatGPT, Cursor or VS Code once, and that chat can use Lakera Guard and every other app you connected to Ace.

Can my own app use Lakera Guard through Ace?

Yes. Create an API key at tryacee.com/developers, then call /api/v1/tools/find and /api/v1/tools/{tool}/run, or use the JavaScript or Python SDK.