arcjet

Implement rate limiting, bot detection, and prompt injection scanning via the Arcjet SDK.

18|Updated Jul 23, 2025
One-click install
npx skills add https://github.com/faw01/create-mf2-app --skill arcjet
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: arcjet
Source: https://github.com/faw01/create-mf2-app/tree/main/apps/cli/template/.agents/skills/arcjet
Command: npx skills add https://github.com/faw01/create-mf2-app --skill arcjet

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill protects your application from automated bot abuse, cost-inflating prompt injections, and unauthorized tool usage by implementing security controls directly within your code.

Core Features & Use Cases

  • Multi-Layer Protection: Provides rate limiting, bot detection, prompt injection scanning, and sensitive information blocking.
  • Flexible Integration: Works across HTTP routes, background jobs, agent tool calls, and MCP servers.
  • Use Case: Secure an AI agent's tool calls by applying a per-user rate limit and a prompt injection scan to ensure the agent only processes safe, authorized requests.

Quick Start

Use the arcjet skill to add rate limiting and prompt injection protection to the weather tool function in my project.

Frequently Asked Questions about arcjet

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I add rate limiting and prompt injection protection to AI agent tool calls?

You can secure AI agent tool calls by applying per-user rate limits and prompt injection scans directly within the tool function code path. This ensures the agent only processes safe, authorized requests.

Can I apply bot detection and rate limiting to background workers and non-HTTP code paths?

Yes, bot detection and rate limiting can be implemented across both HTTP routes and non-HTTP code paths, including background jobs and MCP servers, providing multi-layer abuse prevention throughout your application.

Do I need the Arcjet SDK to implement prompt injection scanning and API protection?

Yes, integrating the Arcjet SDK is required. You must also configure environment-based security keys to activate the abuse prevention controls and sensitive information blocking features.

What is the best way to prevent automated bot abuse and cost-inflating prompt injections?

The best way to prevent automated bot abuse and cost-inflating prompt injections is implementing security controls directly within your code, utilizing rate limiting, bot detection, and prompt injection scanning.

Does rate limiting work with web frameworks to block unauthorized API usage?

Yes, rate limiting works across web frameworks to block unauthorized API usage. It applies configurable limits to HTTP routes, ensuring automated abuse is mitigated before reaching your application logic.

When do I need sensitive information blocking for my application security?

You need sensitive information blocking when securing code paths that process untrusted inputs, such as AI agent tool calls or HTTP routes, preventing unauthorized data exposure alongside standard rate limiting and bot detection.