ArmorIQ Tools
The self-serve developer product at tools.armoriq.ai - a one-line guardrail for Claude Code and OpenAI Codex, plus a console that shows honestly what your coding assistants did.
What ArmorIQ Tools Is
ArmorIQ Tools is the developer-facing product at tools.armoriq.ai: a one-line install that puts a guardrail next to an AI coding assistant, auto-provisions a personal workspace, and gives the developer or their manager an honest record of what the assistant actually did.
Two metaphors the product leans on:
- A seatbelt: it can stop an unsafe action before it happens (Deny or Hold).
- A flight recorder: it shows exactly what happened afterwards (Sessions, session detail, analytics).
It is for individual developers and small teams using Claude Code (via ArmorClaude) or OpenAI Codex (via ArmorCodex) who need to trust, review, and control what those assistants do. The individual installs first; the manager and security lead are who expand and pay.
What You'll Learn
Architecture
How the guardrails, SDK, backend enforce path, and Console fit together, with a system diagram.
The Console
Sessions, session detail, tools catalogue, analytics, AI chat, Policy Studio, and Team.
Enforcement
The four decisions (Allow, Deny, Hold, Error) and how the backend resolves each tool call.
Telemetry
The Honesty Principle: usage read from real records, never invented numbers.
Auth & Identity
Device-code install login, SDK API keys, and Console OAuth with bearer JWTs.
Stack & Deployment
Console, guardrails, SDK, backend, policy engine, and where each is deployed.
System at a glance
The guardrail decides and records; the Console only reads. The guardrail, through the SDK, calls the backend to enforce and to ship telemetry. The Console is a read surface over the evidence the backend has already stored, which is why it can hold to the Honesty Principle.
Tools reuses the same backend (conmap-auto) and the same enforce contract as the ArmorIQ platform. It is a packaging and experience layer, not a separate stack.