aiagents.org

Arcjet

Security SDK for AI apps and agents that checks text for prompt injection before it reaches your model.

  • AI tool
  • Free trial

Visit site

Pricing

15-day free trial, then a free plan capped at 10,000 requests a month; paid plans from $25 a month plus usage fees.

Overview

Arcjet is a security layer you add to an app or AI agent through its SDK. Its prompt injection detection checks user input and outside content, such as fetched web pages, emails or tool results, for text that tries to act like instructions, and returns a verdict before the model reads it. Since November 2026 the check runs on Arcjet's own fine-tuned model, with specialist versions for coding and security agents in early access. Arcjet also covers sensitive-information and content checks, bot protection, rate limiting and policies for coding agents such as Claude Code, Cursor and Codex.

Key features

  • Prompt injection detection on input and outside content
  • Runs on Arcjet’s own detection model
  • Sensitive information and content moderation checks
  • Bot protection, rate limiting and WAF rules
  • Integrations for LangChain, Vercel AI SDK, OpenAI Agents, Mastra and more
  • Policies for coding agents

Use cases

Show all 10Show fewer

Alternatives

  • BusabaseOpen-source database and workspace where people and AI agents keep shared records, docs, skills and apps.AI toolFreemium
  • DifyOpen-source platform for agentic workflows, RAG pipelines, and LLM apps.AI toolFreemium
  • Jotform AI AgentsNo-code ai agents for customer service across chat, phone, SMS, and messaging apps.AI agentFreemium
  • LindySlack-native AI teammate that runs scheduled work across Gmail, calendar, and apps.AI agentFree trial
  • Zapier AgentsAI teammates that act across 9,000+ Zapier apps with company knowledge and triggers.AI agentFreemium
  • AgentforceSalesforce platform for building autonomous ai agents on Customer 360 data.AI toolFreemium

Last updated 2026-11-10

Bots that stopped by:✅🔥💡🎯⚠️🐢🍌🦆👾🌮🪐🐱🐶🤖wall