ArmorClaude
Intent-based security enforcement for Claude Code and Claude Desktop
ArmorClaude
Not to be confused with
ArmorClaude (this section, with an "e") is the plugin for Claude Code and Claude Desktop. ArmorClaw (no "e") is the plugin for OpenClaw chat assistants.
ArmorClaude adds security enforcement to Claude Code and Claude Desktop. It captures Claude's declared intent and applies policy and intent checks before tool execution. Intent requirements depend on the active policy, login state, and tool.
One Command Setup
curl -fsSL https://armoriq.ai/install_armorclaude.sh | bashInstalls the plugin, sets up the ArmorIQ CLI, and optionally connects to the ArmorIQ platform.
What It Does
When you prompt Claude to do something, ArmorClaude:
- Records intent plans for the tools Claude intends to use
- Checks tool calls against policy and, when required, the registered intent plan
- Enforces policy rules set allow/deny rules from any prompt
- Uploads telemetry while signed in, including audit rows, traces, and token usage
Setup
These guides document ArmorClaude 0.2.22. The install command above installs armorclaude@armoriq, uses the production backend, and adds the armoriq CLI.
armoriq login --product armorclaude
armoriq whoamiArmorClaude reads the saved login profile for its product and backend. There is no API-key plugin setting or ARMORIQ_API_KEY fallback. Without a usable login, it starts in monitor mode with intent requirements disabled; local policy rules still apply.
See Configuration for plugin settings and data paths.
Works Everywhere
ArmorClaude is a user-scoped plugin. Once installed, it is active in:
- Claude Code CLI (
claudecommand in terminal) - Claude Desktop app (Code tab, visible under Customize, Personal plugins)
Same enforcement in both. No separate install needed.
Requirements
- Claude Code 2.x or Claude Desktop
- Node.js 20.19 or newer, npm, Git, and
curl - macOS, Linux, or Windows with WSL or Git Bash
Stuck?
See the troubleshooting page in this section, or email license@armoriq.io.