Skip to main content

Waxell Endpoints

Endpoints is the part of Waxell that governs the AI tools your team runs locally — the desktop apps, IDEs, CLIs, and browser tabs on real laptops, not just the agents you build with the Waxell SDK. It answers three questions every security and platform team is now asking:

  1. What AI is actually running on our machines? (shadow-AI discovery)
  2. What is it doing? (per-app, per-device, per-user activity)
  3. Can we control it? (block, allow, or capture — under policy)

The problem: ungoverned AI on the endpoint​

AI tools install themselves one developer at a time. Claude Desktop, Cursor, Claude Code, GitHub Copilot, ChatGPT's desktop app, Perplexity, a dozen browser-based assistants — each one talks to a model provider over the network, often touching source code, customer data, or secrets. Most organizations have no inventory of which of these are running, on whose machine, against which provider.

Waxell Endpoints closes that gap with a small, signed agent you deploy to a whole fleet (through your MDM or RMM) or to a single machine. Once installed, it:

  • Discovers every AI app on the machine and reports it to your workspace.
  • Captures AI chats in the browser through the Waxell browser extension — the prompt, the reply, the model and the account used — on ChatGPT, Claude, Gemini, Copilot, Perplexity and any site you add.
  • Governs coding agents — Claude Code, Codex, Cursor, Gemini CLI, Copilot, Claude Desktop and others — through hooks that check each command, file access and prompt before it runs.
  • Contains AI use when policy says so: block sites, vendors or apps, bar users, and quarantine a device after a serious violation.

What is captured, and the privacy model​

Browser capture is on by default

Installing the agent and extension turns on browser AI capture for the built-in AI sites. Every capture is redacted on the device — API keys, tokens, private keys, card numbers, SSNs and other secrets are replaced before anything leaves the machine. Turn Browser capture off on the Guard to hold captures on the device instead; you can do that for the whole workspace, a team or a single device.

Waxell does not decrypt your network traffic. Browser AI is read inside the page by the extension, and coding agents report through their own hooks. Non-AI browsing — banking, mail, everything else — is never captured. See Privacy & DLP.

Governed vs. ungoverned: the gap analysis​

The product is, at heart, a gap analysis. For every machine it shows you:

Ungoverned (before)Governed (after Waxell)
InventoryUnknown which AI apps existEvery AI app, per device, listed
AttributionNo idea who/what is calling providersPer-process, per-user, per-host flows
ControlNoneAllow, ask, block, or contain under policy
EvidenceNoneRedacted, attributed activity record

Two ways to deploy​

  • A whole fleet → MDM. Push the signed installer and your workspace's profiles from Intune, Jamf, Kandji, Hexnode or an RMM such as NinjaOne. Devices enroll silently, with no end-user action.
  • One machine → Manual install. Download the signed app and install it yourself in a couple of minutes — no MDM, no IT. Great for trying Waxell on your own laptop first.

Where to go next​