Arcjet logo

Arcjet

Secure the AI agents you're building at runtime

Built by Shane O'Connor, Cassie Gatton, David Mytton and 1 other · Paid

Visit website

Arcjet is a Product Hunt listing on Launch Llama with 430 total upvotes, founded by Shane O'Connor, Cassie Gatton, David Mytton and 1 other who joined in Sep 2026, compared against 6 alternatives.

Product Hunt listing

Listing status

430upvotes

Community score

Shane O'Connor, Cassie Gatton, David Mytton and 1 other (Joined Sep 2026)

Founder

Launched on Launch Llama Sep 27, 2026 · #2 of 47 that day

Arcjet

About

Arcjet is the runtime security platform that ships in your AI code. Detect prompt injection, authorize agent tool calls, redact sensitive data, and block bots and abuse. Real-time security building blocks you call inside your app, before an action happens.

Competitors: Predibase, Robust Intelligence, AI Verify, Lakera AI, Rebuff

Launch Llama's Take

Based on Launch Llama data as of

Arcjet has 430 upvotes, including the votes it received on Product Hunt. It has been listed on Launch Llama since September 2026. 1 Launch Llama member has upvoted it and it has no reviews yet.

Arcjet is a runtime security platform that provides building blocks for detecting prompt injection, authorizing agent tool calls, and redacting sensitive data. Compared with Kastra, which is described as a runtime authorization layer for AI agents, Arcjet's listed features also include blocking bots and abuse on web applications.

Arcjet vs Kastra

Arcjet is a runtime security platform that provides building blocks for detecting prompt injection, authorizing agent tool calls, and redacting sensitive data. Compared with Kastra, which is described as a runtime authorization layer for AI agents, Arcjet's listed features also include blocking bots and abuse on web applications.

Arcjet vs Kastra: side-by-side comparison
ArcjetKastraArceris Ai Security GatewayJet Admin
What it doesSecure the AI agents you're building at runtimeFor AI teams: Kastra enforces authorization policies on Claude, Cursor, and OpenAI agents in sub-1ms so unauthorized tool use and prompt…AI security gateway that masks PII in real timeFor operations and support teams: Jet Admin builds AI agents without code that take action across 200+ tools via Slack, WhatsApp, or Tele…
PricingPaidFreemiumPaidFreemium
Free planNo free plan listedYes, free planNo free plan listedYes, free plan
CategorySecurity softwareDeveloper ToolsAutomation toolsDeveloper Tools
Upvotes on Launch Llama430398Not stated316
Listed on Launch LlamaSeptember 2026July 2026November 2026May 2026

See all Arcjet alternatives

Key Features

  • Detect prompt injection in user input and tool outputs
  • Authorize agent tool calls based on identity and role
  • Redact sensitive data before model context or third party calls
  • Control token spending with per-user budget limits
  • Block bots and abuse on web applications

Use Cases

  • Integrate Arcjet's prompt injection detection directly into your agent code to block malicious inputs before they reach your LLM
  • Use Arcjet's tool call authorization to verify which agent actions are allowed before executing them against your backend systems
  • Call Arcjet's data redaction API within your agent workflows to automatically strip sensitive information before logging or storage
  • Deploy Arcjet's bot and abuse detection at runtime to identify and block suspicious requests targeting your AI agent endpoints
  • Embed Arcjet security blocks into your agent initialization to enforce policies across all tool calls and external integrations

Who it is for

  • Engineering teams building and shipping AI agents into production workflows
  • Security teams needing visibility and policy enforcement over coding agents like Claude Code, Cursor, and Copilot
  • Developers adding runtime protection to web apps and APIs using supported SDKs such as Next.js, Express, or FastAPI

Pricing

Individual plan costs $25/month. Startup plan costs $299/month. Enterprise available at custom pricing. Usage fees apply at $5 per 1M web requests and $50 per 1M agent requests. 15-day free trial available.

View Arcjet pricing →

FAQ

Can I book a demo?

Of course! Book a demo with our team.

How does usage-based pricing work?

Any protect() SDK call counts as a web request, billed at $5 per million. Rate limit, filter, and Shield WAF rules are included in that single request. Protecting coding agents and custom agents counts as an agent request, billed at $50 per million.

Are there any request limits?

There are no hard request limits on paid accounts (the trial has usage limits). All protections are billed based on actual usage at the per-unit rates shown in the usage fees table. You only pay for what you use and you can set budget thresholds to avoid surprises. Billing guarantees are available to insure against volumetric attacks - your app will never go down because you hit a limit.

Is there a free plan?

Yes. When the 15 day trial ends, your account continues on a free plan limited to 10,000 requests per month. The limit is a hard stop rather than metered overage, so you are never billed for going over it. Set up billing and choose a plan to lift the cap.

Do you support open source projects?

Yes. Many open source projects use Arcjet on the Individual plan. The Individual plan works well for non-commercial, community-maintained open source projects and includes access to all protections with discounted or free pricing. Reach out to us to discuss how Arcjet can help.

What payment methods do you accept?

We accept all major credit and debit cards on all plans. Enterprise plans can optionally be purchased through an annual contract and paid by ACH/Wire.

Who is behind Arcjet?

Arcjet Labs, Inc. is headquartered in San Francisco and also has an office in New York. We raised a seed round in 2023 and Series A in 2025. We're backed by Andreessen Horowitz, Plural, Seedcamp, and 20+ angel investors. Our CEO, David Mytton, writes the console.dev devtools newsletter.

Alternatives

Arceris Ai Security GatewayProtect sensitive data and control what AI agents can access and doJet AdminFor operations and support teams: Jet Admin builds AI agents without code that take action across 200+ tools via Slack, WhatsApp, or Telegram.KastraFor AI teams: Kastra enforces authorization policies on Claude, Cursor, and OpenAI agents in sub-1ms so unauthorized tool use and prompt injection never reach production.ClawSecureFor AI developers: ClawSecure scans and monitors AI agents in real-time with full OWASP ASI coverage, sub-200ms verification, and no signup required.AI Security GuardFor AI teams: AI Security Guard blocks jailbreaks and prompt injection attacks in real time so your agents stay protected from malicious inputs.AiruncodeFor developers: Airuncode runs multiple local coding agents on your machine so you ship faster with self-healing code and zero markup.

New on Launch Llama

Launched this week · This month's launches · All products A–Z

Tags

#prompt injection#ai agents#tool authorization#sensitive data redaction#token budgets#bot protection#runtime security#llm guardrails

Ask AI about Arcjet

ChatGPT Claude Perplexity Grok

Supporters

1 founder

1 founder supported Arcjet on Launch Llama.

Andreas Christodoulou

Third-party sources

For bots & AI agents

Machine-readable versions of Arcjet and Launch Llama. Fetch these instead of parsing this page.

This listing (.md) llms.txt llms-full.txt ai.txt Full catalog (.md) Endpoint index API docs MCP server MCP manifest Developer docs