Your GCP API keys have a kill switch.
One that can't be turned against you.

That's exactly the right question to ask. Most tools request broad OAuth access or Owner-level permissions. CloudSentinel requests a Custom IAM Role with a permission surface of exactly three things: read your API key list, read your usage metrics, and delete a key. The delete permission only works on existing keys — the role does not contain apikeys.create. This isn't a policy we promise to follow. It is a structural absence at the GCP permission level. Even if our entire platform was compromised, an attacker could only remove keys from your project — never create infrastructure, never access your data, never touch anything else.

No. CloudSentinel sits entirely outside your critical path. Your API keys work independently of our platform. If we go offline, your keys keep functioning exactly as they did before — we simply won't be monitoring them during that window. The Alerting Policies we create inside your GCP project are owned by your project, not ours. The only thing that stops working is the auto-revocation. We maintain 99.9% uptime SLA on the webhook receiver specifically because it is the one critical function.

No. The GCP IAM permissions we hold allow us to see that a key exists and its usage metrics — request counts and billing data. We cannot see the key's actual secret value after creation, and we have zero visibility into what your API keys are being used for or what data flows through them. We see a number go up. That is all.

You will receive an immediate alert when a key is revoked, including the exact metric that triggered it. You can then issue a new key from your GCP console and adjust the threshold. We are working on a one-click temporary suspension mode that pauses monitoring without revoking — this is on the Phase 2 roadmap. For now we recommend setting thresholds conservatively above your highest known peak traffic, with at least a 3x safety margin.

GCP budget alerts send you an email. They do not revoke keys. By the time you read the email, open the console, navigate to API keys, and delete the offending key — assuming you even know which one it is — significant damage may already be done. CloudSentinel receives the Pub/Sub alert programmatically and fires the DeleteKey command automatically — no human in the loop, no manual login, no finding which key caused it. GCP Monitoring evaluates thresholds on a 1-5 minute cycle. The gap between breach and revocation goes from hours (waiting for a human) to minutes (fully automated). While you're still asleep, the key is already dead.

We store the minimum required to operate: your GCP project ID, the key IDs you are monitoring, the thresholds you set, and a log of revocation events for your audit trail. We do not store usage metrics — those live in GCP. We do not store key values — those are never accessible to us. All stored data is encrypted at rest. You can delete your account and all associated data at any time.

This is exactly why we don't rely on spend or billing data at all. GCP billing metrics can be delayed by several hours — by the time your budget alert fires, the damage is already done. CloudSentinel works entirely on raw API request count, which GCP updates in near real-time. The moment your request threshold is crossed, Google fires an alert and CloudSentinel revokes the key automatically — within minutes, before your billing data has even registered the spike.

Starter at $19/mo covers up to 3 GCP projects and 10 API keys. Pro at $49/mo covers 10 projects and unlimited keys. Business at $99/mo is unlimited everything with Slack and PagerDuty alerts. Early access members lock in 50% off their chosen plan for the first 3 months. There is no free tier — this is a professional security tool and we price it accordingly.