Is it safe to put company code into Cursor or Copilot?

Not by default. Whatever you paste or open gets sent to the AI provider's servers, including secrets and customer data. Emil sits between your tools and the model and catches API keys, credentials, customer data, and source code before they ever leave.

Can AI coding tools see my API keys and .env files?

Yes, whenever they end up in the files or prompts you send, which is the most common way secrets leak. Emil detects keys, tokens, passwords, and .env values and redacts them before the prompt reaches the model.

Does Emil slow down my AI?

No. Emil screens in the moment and hands the model a clean prompt, so you get the same responses at the same speed.

Does Emil store my prompts or my code?

No. Emil processes your text in memory and keeps only a short screening record (the decision, risk score, and a redacted preview), never the full content. Classification runs on Emil's own infrastructure, is never sent to a third-party model, and is never used for training.

Which tools does Emil work with?

Anything OpenAI-compatible: Cursor, Claude Code, Copilot, Lovable, v0, ChatGPT, Gemini, DeepSeek, Ollama, and self-hosted models. Point your tool at Emil instead of the provider, or use the browser extension if you don't write code.

Do I need to be technical to use it?

No. Developers change one line of config (the base URL). Everyone else uses the one-click browser extension that screens prompts before they reach the AI.

AI Builders

The guardrail layer for your AI product

Stop your AI from generating things it shouldn't. Emil screens every prompt and response for unsafe content, PII, prompt injection, and off-brand output — point your base URL at Emil and ship.

Protect your team For developers

The problem

Your AI can be jailbroken into generating explicit, harmful, or off-brand content that lands on your users and your liability.
Users paste sensitive data into your product, putting it in your logs and your AI provider's.
You need to prove to enterprise buyers that your AI is safe — with evidence, not promises.

What Emil catches for ai builders

Explicit, harmful, and unsafe output (14 safety categories)
Prompt-injection and jailbreak attempts
PII and secrets in prompts and responses
Custom brand and policy rules you define in natural language

Regulations this maps to

EU AI Act
App-store & platform safety policies
Customer DPAs

A content filter reduces regulatory risk but isn't compliance on its own. Review presets with counsel.

Questions

How do I add Emil to my AI product?

Change one line: point your OpenAI-compatible client's base URL at Emil. Every request and response is then screened — no SDK, no middleware. Anthropic-compatible and self-hosted models are supported too.

Can Emil stop my AI generating explicit or unsafe content?

Yes. Emil screens the model's output across 14 safety categories with a deterministic explicit backstop plus a model classifier, and blocks or redacts before the response reaches your user.

Can I define my own rules?

Yes. Beyond the built-in policies, you can write natural-language custom rules — 'never compare us to competitors', 'never give medical advice' — and Emil enforces them on output.

Can I prove the guardrails work?

Emil ships a red-team evaluation harness that scores your policy's catch rate across unsafe categories and reports a false-positive rate — the evidence enterprise buyers and auditors ask for.

Other industries

Financial Services Customer Support AI Education