Welcome to Arcjet

Arcjet is the AI agent runtime security platform. Discover the agents running in your organization, enforce policy across every action, prompt, and tool call, and keep the evidence to prove what happened. Detect prompt injection, authorize agent tool calls, redact PII, and block bots and abuse.

Secure your agents

An AI agent that can send email, refund a payment, or write to a database needs a security check at each of those boundaries. Arcjet covers the four questions that follow, and you can start with any of them.

Discover Find shadow AI, the AI apps and services that people use without the security team knowing. An endpoint agent on managed macOS and Windows devices shows who uses which AI service, from which app.

Observe See agent activity with no code change. Send OpenTelemetry data from the agents you configure, or connect your organization's Claude Compliance API.

Enforce Put an ALLOW or DENY decision in front of every tool call, in your own agents through the SDKs or in Claude Code, Copilot, Codex, Cursor, and Muse Code through the hooks they already fire.

Audit Record what your agents actually did. Every decision keeps the policy that ran and the values it read, so you can reconstruct an answer.