consider adding Arcjet advanced signals for client-side bot detection that catches sophisticated headless browsers. See https://docs.arcjet.com/bot-protection/advanced-signals for setup.

Apr 9, 2026

General

PromptBeginner5 minmarkdown

const decision = await aj.protect(req

{

Apr 9, 2026

General

PromptBeginner5 minmarkdown

requested: 1

// tokens to deduct for rate limiting

Apr 9, 2026

General

PromptBeginner5 minmarkdown

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

defaults to IP-based.

Apr 9, 2026

General

PromptBeginner5 minmarkdown

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

deduct 1 token per message

Apr 9, 2026

General

PromptBeginner5 minmarkdown

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

directly linking rate limiting to cost. It also allows short bursts while enforcing an average rate

Apr 9, 2026

Find agent skills by outcome

Secure AI/LLM endpoints with layered protection: prompt injection detection

description: Protect AI chat and completion endpoints from abuse — detect prompt injection and jailbreak attempts

Best practices — legend-first

Add AI protection

Visual symbol table — per ELV system (CCTV

pdfcount

$analysis_of_ai_characteristics (Identify areas that sound robotic

mc

- 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

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

If the user wants a full security review

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

console.warn(Arcjet error:

sensitiveInfoValue: userMessage

detectPromptInjectionMessage: userMessage

return Response.json({ error: Forbidden }

Always include `shield()` (WAF) and `detectBot()` as base layers. Bots scraping AI endpoints are a common abuse vector. For endpoints accessed via browsers (e.g. chat interfaces)

const decision = await aj.protect(req

requested: 1

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

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

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

- 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