Skip to main content

​
What this covers

AI-assisted IDEs and coding assistants where the LLM provider and orchestration are hosted by the vendor — you can’t intercept the call server-side. Enforcement happens on the developer’s managed device. This includes:

​
Architecture

developer's managed device
    │
    ├── native IDE / CLI ──► PAC ──► TrustGate Endpoint (MITM) ──► LLM provider
    └── Chromium browser ──► Browser extension ──► LLM provider web app
The developer does not install TrustGate software. Both integrations are pushed by MDM.

​
Prerequisites

  • An Endpoint integration — provides the PAC file URL, the client certificate+key, and the TrustGate CA certificate used for dynamic TLS.
  • A Browser integration — provides the MDM-distributable extension package for Chromium-based browsers.
  • An MDM tool (Jamf, Intune, Kandji, Workspace ONE, or similar) with enrollment on developer machines.
  • Chromium-based browsers only for the Browser surface (Chrome, Edge, Brave, Arc, Opera). Firefox and Safari are not supported.

​
Wire it up

​
Endpoint — native IDE traffic

  1. Create an Endpoint integration in the console. Download the PAC URL, the per-org client certificate + key, and the TrustGate CA certificate.
  2. Using your MDM, push to managed developer machines:
    • The TrustGate CA certificate into the system trust store (so dynamic TLS certificates are trusted by native apps).
    • The client certificate + key into the system keychain (used for mTLS to identify your org).
    • A proxy configuration that uses the PAC URL so the OS-level HTTPS traffic for the targeted IDEs is routed through the TrustGate Endpoint MITM.
  3. In TrustGate, configure the Endpoint policies to match the LLM endpoints those IDEs call (for example, api.openai.com, api.anthropic.com, api.githubcopilot.com, api.cursor.sh). Anything outside the scope defined on the integration is not intercepted.
Developers do not sign in, install software, or change settings.

​
Browser — web AI apps

  1. Create a Browser integration. Download the MDM extension package.
  2. Push the extension to Chromium browsers via MDM force-install policies (for Chrome and Edge, use their enterprise policy for ExtensionInstallForcelist).
  3. Configure which AI chat apps the extension activates on (for example, chatgpt.com, claude.ai, gemini.google.com, cursor.com, github.com/copilot).
The extension is bound to your tenant via the managed extension configuration — developers do not authenticate.

​
Verify

  1. On a test managed machine, open Cursor or use GitHub Copilot in an IDE. Send a prompt.
  2. Open Runtime → Explorer and filter by the Endpoint integration. The prompt and response should appear.
  3. Separately, open chatgpt.com in Chrome on the same machine and send a prompt. It should appear filtered by the Browser integration.

​
Policies to turn on first

  • Source-code IP leakage — pattern and semantic detectors for proprietary identifiers, internal project names, and secret tokens.
  • Data protection & masking — PII and credentials embedded in code or pasted prompts.
  • Prompt security — jailbreak attempts originating from pasted content (indirect prompt injection).
  • Keyword block — list of project codenames or customer identifiers that must not leave the device.

​
Limitations

  • Streaming on Endpoint: many AI IDE providers use streaming responses. Mask and Block actions apply to the final consolidated payload; the detector decision is enforced once the stream completes.
  • Certificate pinning: some native vendors pin their API certificates and will reject the TrustGate dynamic TLS certificate. Consult the vendor’s enterprise documentation; a few IDEs offer an MDM-managed CA override.
  • Browser scope: only Chromium-based browsers are supported. Firefox and Safari are not covered.
  • Unmanaged devices: BYOD laptops without MDM are out of scope for both surfaces. Restrict access from unmanaged devices via SSO / conditional access upstream.
  • Internal chat apps: these surfaces protect your developers when they use the approved public LLM tools. Homegrown internal chat apps should be covered with the Gateway or API surfaces, not Browser or Endpoint.