Getting Started

Quick Start

Install ArmorClaude and see it enforce tool calls in under 2 minutes

Quick Start

Get ArmorClaude running in under 2 minutes. An ArmorIQ account is required to authenticate the plugin.

Prerequisites

  • Claude Code 2.x or Claude Desktop
  • curl (preinstalled on every macOS + Linux)
  • macOS, Linux, or Windows with WSL or any bash-compatible shell (Git Bash works; Git itself is not required)

Install

curl -fsSL https://armoriq.ai/install_armorclaude.sh | bash
curl -fsSL https://armoriq.ai/install_armorclaude.sh | bash

Run from WSL or Git Bash, not PowerShell/CMD.

The installer handles everything: adds the marketplace, installs the plugin, installs the ArmorIQ CLI, and optionally connects your ArmorIQ account.

When prompted "Connect your ArmorIQ account now?" press Y to authenticate via browser. Declining the prompt aborts the install. You can re-run the installer or armoriq login anytime to connect.

Manual install (without the curl script):

claude plugin marketplace add armoriq/armorClaude
claude plugin install armorclaude@armoriq

Verify Installation

After the installer finishes, verify everything is set up:

claude plugin list

You should see:

Installed plugins:

  armorclaude@armoriq
    Version: 0.2.4
    Scope: user
    Status: enabled

If you connected your ArmorIQ account during install, verify the login:

armoriq whoami

You should see:

ArmorIQ Credentials

  Email:    you@company.com
  API Key:  ak_live_xxxxxxxx...
  User ID:  ...
  Org ID:   ...
  File:     ~/.armoriq/credentials.json

If you skipped the login, you can connect anytime:

armoriq login

This opens your browser. Sign in with Google or GitHub, click Authorize, and the key is saved automatically. ArmorClaude picks it up on the next Claude session.

Try It

Open Claude Code in any project:

mkdir -p /tmp/demo && cd /tmp/demo
echo "# My Project" > README.md
claude

Prompt:

Read README.md and tell me what's in it.

What happens:

  1. Claude calls register_intent_plan declaring it will use Read
  2. ArmorClaude stores the plan
  3. Claude calls Read. ArmorClaude checks it against the plan and allows it
  4. Claude returns the file contents

You will see in the transcript:

register_intent_plan: Intent registered: 1 steps.
Read 1 file

Block Something

Set a policy rule using the /armor command:

/armor policy add deny WebFetch

ArmorClaude stages the change and shows a diff. Confirm it:

/armor yes

Now try:

Fetch https://example.com

Result:

ArmorClaude policy deny: stmt_1

The tool call is blocked before execution. Claude sees the denial and tells the user.

To remove the rule:

/armor policy remove stmt_1
/armor yes

What You Just Saw

FeatureWhat happened
Intent planClaude declared Read before using it
Plan enforcementRead was in the plan, so it was allowed
Policy ruledeny WebFetch blocked the fetch call
Policy managementRules created and deleted from the chat prompt

Next Steps

On this page