set one up first with `shield()` as the base rule. The user will need to register for an Arcjet account at https://app.arcjet.com then use the `ARCJET_KEY` in their environment variables.

Apr 10, 2026

General

PromptBeginner5 minmarkdown

Detects jailbreaks

role-play escapes

Apr 10, 2026

General

PromptBeginner5 minmarkdown

Read https://docs.arcjet.com/llms.txt for comprehensive SDK documentation covering all frameworks

rule types

Apr 10, 2026

Find agent skills by outcome

emotion: Fierce

Viking

prompt: You will perform an image edit using the person from the provided photo as the main subject. The face must remain clear and unaltered. Transform the subject into a steely-eyed **Wild West Gunslinger/Outlaw**

location: The wooden prow of a carved dragon-headed longship

- Sensitive info detection runs **locally in WASM** — no user data is sent to external services. It is only available in route handlers

**Vercel AI SDK**: Arcjet works alongside the Vercel AI SDK. Call `protect()` before `streamText()` / `generateText()`. If denied

**Multiple models / providers**: Use the same Arcjet instance regardless of which AI provider you use. Arcjet operates at the HTTP layer

If the user wants a full security review

Adapt the response format to your framework (e.g.

- `list-requests` — confirm decisions are being recorded

detectPromptInjectionMessage: userMessage

return Response.json({ error: Forbidden }

console.warn(Arcjet error:

sensitiveInfoValue: userMessage

requested: 1

const decision = await aj.protect(req

Pass the `requested` parameter at `protect()` time to deduct tokens proportional to model cost. For example

Set `characteristics` to track per-user: `[userId]` if authenticated

Use `tokenBucket()` / `token_bucket()` for AI endpoints — the `requested` parameter can be set proportional to actual model token usage

- JS: `sensitiveInfo({ deny: [EMAIL

Arcjet rules run **before** the request reaches your AI model — blocking prompt injection

Check for an existing shared Arcjet client (see `/arcjet:protect-route` for full setup). If none exists

Detects jailbreaks

Read https://docs.arcjet.com/llms.txt for comprehensive SDK documentation covering all frameworks

prompt: You will perform an image edit using the person from the provided photo as the main subject. The face must remain clear and unaltered. Transform the subject into a steely-eyed Wild West Gunslinger/Outlaw

- Sensitive info detection runs locally in WASM — no user data is sent to external services. It is only available in route handlers

Vercel AI SDK: Arcjet works alongside the Vercel AI SDK. Call `protect()` before `streamText()` / `generateText()`. If denied

Multiple models / providers: Use the same Arcjet instance regardless of which AI provider you use. Arcjet operates at the HTTP layer

Arcjet rules run before the request reaches your AI model — blocking prompt injection