# ArmorIQ - AI Agent Security Platform

> ArmorIQ delivers intent-aware visibility, cryptographic trust, and continuous
> policy enforcement for enterprise AI. It's not about Identity, it's about Intent.
> ArmorIQ sits between an agent's reasoning and its execution: agents plan freely,
> but every action is verified against the approved task before it runs.

## Platform Products
- [Gatekeeper](https://armoriq.ai/products/gatekeeper): Enforce cryptographic access policies on MCP endpoints. Gatekeeper blocks unauthorized agent actions before they execute with zero-trust enforcement.
- [Intent Engine](https://armoriq.ai/products/intent-engine): Classify and verify AI agent intents in real-time. The Intent Engine ensures every autonomous action aligns with approved business logic before execution.
- [Registry](https://armoriq.ai/products/registry): Discover, catalog, and monitor all MCP endpoints across your organization. The Registry provides a single source of truth for your AI agent infrastructure.
- [Sentry](https://armoriq.ai/products/sentry): Scan MCP endpoints for vulnerabilities, misconfigurations, and compliance gaps. Sentry provides continuous security monitoring for AI agent infrastructure.
- [Auditor](https://armoriq.ai/products/auditor): Full audit trails for every AI agent action. Auditor provides compliance reporting, forensic analysis, and real-time alerting for autonomous agent operations.

## Developer Tools
- [ArmorClaude](https://armoriq.ai/tools/armorclaude): Drop intent enforcement into Claude Code in one line. ArmorClaude binds every tool call to a signed intent plan, blocking intent drift and unauthorized actions before they execute.
- [ArmorCodex](https://armoriq.ai/tools/armorcodex): Bash gating for OpenAI Codex with no code changes required. ArmorCodex binds every Bash command to a signed intent plan, gating execution before it runs.
- [ArmorCopilot](https://armoriq.ai/tools/armorcopilot): ArmorCopilot adds intent assurance to GitHub Copilot. Every action is verified against your policies before it runs. No drift, no escalation, no unauthorized behavior.
- [ArmorClaw](https://armoriq.ai/tools/armorclaw): ArmorClaw adds intent assurance to your OpenClaw agents. Every action is verified against your policies before it runs. No drift, no escalation, no unauthorized behavior.

## Key Concepts
- Intent enforcement: controlling what an agent does once it has access, not just who has access (which is IAM's job).
- MCP security: scanning and governing Model Context Protocol endpoints used by AI agents.
- Intent assurance: cryptographically binding every agent action to a signed intent plan, fail-closed.

## Who it's for
Security teams, platform engineers, and compliance officers running AI agents in production.

## Navigation
- [Homepage](https://armoriq.ai/)
- [About Us](https://armoriq.ai/about-us)
- [Manifesto](https://armoriq.ai/manifesto)
- [FAQ](https://armoriq.ai/faq)
- [Blog](https://armoriq.ai/blogs)
- [Articles](https://armoriq.ai/articles)
- [Events](https://armoriq.ai/events)
- [Videos](https://armoriq.ai/videos)

## Contact
For enterprise inquiries and demos, visit https://armoriq.ai/book-a-demo