What you’re testing. Shadow AI runs on your Mac and inspects the traffic
going to AI tools (Claude, ChatGPT, and others) on your device, before
anything is sent. When it sees a secret or sensitive data, it blocks or
masks it. It’s the real product, so it inspects your system-wide AI
traffic while it’s on, and it’s fully reversible (turn it off, or uninstall,
any time).
Before you start
A Mac on macOS 13 or newer
The app also runs on macOS 12, but there it uses the system proxy rather
than the network extension, and fewer apps are captured.
A PromptGuard account
Any plan works: personal Shadow AI is included with every plan. Don’t have
an account? Sign up free at
app.promptguard.co.
Step 1: Install
1
Download the app
Go to promptguard.co/download and
click Download for macOS. It’s a signed, notarized Apple build
(
PromptGuard.Shadow_<version>_universal.dmg), so macOS opens it without
warnings.2
Drag to Applications
Double-click the
.dmg, then drag PromptGuard Shadow into your
Applications folder.3
Launch it
Open PromptGuard Shadow from Applications. A shield icon appears in
your menu bar (top-right of the screen); Open PromptGuard in its menu
opens the app window. Keep the app running: it checks AI traffic only
while it is open.
Step 2: Sign in
1
Connect this device
In the app window, click Connect this device. Your browser opens to the
PromptGuard sign-in page.
2
Sign in with your account
Log in with your PromptGuard account (or create one on the spot). The
browser hands you back to the app automatically, nothing to copy or paste.
Step 3: Turn on protection
1
Click “Turn on protection”
In the app window, click Turn on protection.
2
Allow the network extension, once
macOS says PromptGuard wants to add a network extension. The app shows a
button that opens the right page of System Settings: allow
PromptGuard Shadow there. Until you do, nothing is checked. This is how
the agent sees AI traffic from every app, including command-line tools,
without changing your proxy settings.
3
Approve the certificate, once
macOS will ask you to approve a secure-inspection certificate and prompt
for your Mac password. This is expected and required: it’s how the agent
reads encrypted AI traffic to inspect it. Every tool of this kind (corporate
VPNs, DLP) asks for it. You’ll only be asked once.
4
Confirm it's on
The app should now read “You’re protected.” Leave it running in the
background and go use Claude as you normally would.
Step 4: Try to leak a secret (the fun part)
This is what you’re really here to test. Open claude.ai and paste each of these into a new chat, one at a time. Watch what happens.Try your own realistic examples too, a fake password, a customer email, a
snippet of code with a hard-coded token. The more real-world prompts you throw
at it, the more useful your feedback.
Step 5: See every decision
Open the app and go to Activity. You’ll see a live log of every inspected request: what tool it was going to, the decision (allowed / redacted / blocked), why, and a masked preview (the raw secret is never stored or shown). This is the audit trail an admin would see across a whole fleet.What good looks like
- Safe prompts pass through instantly, you shouldn’t feel the agent.
- Secrets (API keys, tokens, passwords) get blocked; PII (phones, emails) gets redacted.
- The Activity log shows each catch with a clear reason.
- Claude keeps working normally the whole time.
Giving feedback
We want the rough edges. As you test, please note and send back:False negatives
Something sensitive that got through un-caught. The most valuable
feedback, tell us exactly what you typed and where.
False positives
A safe prompt that got blocked or mangled when it shouldn’t have been.
Friction
Anything confusing, slow, or annoying, install, sign-in, the extension and
certificate prompts, the app, performance.
Breakage
Any AI tool or website that stopped working, or connectivity issues while
protection was on.
Turning it off & uninstalling
You’re always in control:1
Pause anytime
In the shield menu, choose Turn protection off. AI traffic flows normally
again; the app stays installed.
2
Remove completely
In the app, Settings → Remove PromptGuard. This removes the network
extension, removes the inspection certificate, and deletes local data,
leaving nothing behind. Then drag the app to the Trash.
Good to know
Will this slow down my Mac or break other apps?
Will this slow down my Mac or break other apps?
No. The agent only decrypts and inspects traffic to the AI tools it
monitors; everything else, email, Slack, streaming, your VPN, is not
decrypted. (The network extension also hands connections to
google.com
to the agent, which passes them on without decrypting them.) Safe AI prompts pass
quickly.I see a certificate warning in a browser tab
I see a certificate warning in a browser tab
If you turned on protection while a tab was mid-load, refresh the tab. If a
warning persists on an AI site, that’s worth reporting, include the site.
Google / some sites feel occasionally flaky
Google / some sites feel occasionally flaky
Some sites use a newer transport (HTTP/3 over QUIC). The network extension
refuses QUIC to AI sites, so the browser falls back to HTTPS, and you may
see a brief hiccup the first time. See
Known Limitations. Report it if
a site is unusable.
Is my data being sent anywhere?
Is my data being sent anywhere?
Prompts are read on your device, masked there, and the masked prompt is
sent to the PromptGuard engine for a verdict. Only a masked preview
(never the raw secret) is stored locally for the Activity log. See
Privacy & data handling.
Next steps
Fleet enrollment
Roll Shadow AI out to every device
Customise policies
Tune checks, warn instead of block, add your own rules
Hit a wall?
The troubleshooting guide covers the common install and connectivity issues,
or just email support@promptguard.co and we’ll jump on it.