Quick Start Guide

Welcome to Probecast! This guide will help you get started in just a few minutes.

Step 1: Create Your Account

Pick a plan (there's a free one) at probecast.io/pricing and create your account. You can use your email or sign in with Google.

Step 2: Add APIs to Monitor

Once logged in, you have two options:

• Browse Public Catalog: Choose from 1,000+ pre-configured APIs (Stripe, SendGrid, AWS, etc)
• Add Custom APIs: Enter your own API endpoints — on paid plans (Indie and up); Startup and Agency check from all 4 regions

Step 3: Configure Alerts

Go to Settings → Alerts and choose how you want to be notified:

• Email notifications (all plans)
• Slack notifications (all plans)
• Web push notifications in the browser (all plans)
• Webhook integrations (Startup & Agency)

Step 4: View Your Dashboard

Your dashboard shows real-time status of all monitored APIs. You'll see:

• Current status (up/down/degraded)
• Uptime percentage (24h, 7d, 30d)
• Average latency from 4 global regions
• Recent incidents

Setting Up Alerts

Stay informed instantly when something goes wrong. Probecast offers multiple notification channels.

Alert Types

Adding APIs to Monitor

From Public Catalog

1. Click "Browse APIs" in your dashboard
2. Use search or filter by category (Payment, Email, SMS, etc)
3. Click the "+" icon on any API card
4. Confirm to add it to your dashboard

Custom APIs (paid plans)

Monitor your own APIs and internal endpoints:

1. Go to Dashboard → Custom APIs → Add New
2. Fill in the information:
   • Name: Friendly name (e.g., "Payment API - Production")
   • URL: The endpoint to monitor
   • Method: GET, POST, PUT, DELETE or HEAD
   • Expected Status: HTTP codes considered healthy (typically 200-299)
   • Headers: (Optional) Authentication or custom headers
   • Body: (Optional) Request body for POST/PUT
3. Click "Start Monitoring"

Limits per Plan

Understanding Status Codes

Probecast categorizes API health into three states:

● Up (Healthy)

The API is responding correctly with expected status codes. All systems normal.

● Degraded

The API is responding but slower than usual or intermittently failing. May indicate issues.

● Down (Unhealthy)

The API is not responding, returning error codes or timing out. Action required.

How We Calculate Uptime

Uptime is calculated as:

Check Frequency

We check each API every 5 minutes from 4 geographic regions:

• 🇺🇸US Westus-west
• 🇺🇸US Eastus-east
• 🇸🇬Asiaasia
• 🇪🇺Europeeurope

Example

If an API had 285 successful checks out of 288 total checks in 24 hours:

Multi-Region Monitoring

Probecast monitors APIs from 4 strategic global locations to ensure comprehensive coverage:

How It Works for Customers

Understand the complete monitoring and notification flow of Probecast.

Applications

Applications let you group monitors by project. Instead of a flat list of APIs, you organize monitors around your actual products — each Application gets its own health score, status page, alert rules and dashboard view.

Why Applications?

An agency looks after several client sites; a founder runs several products. Either way, one Application per client or product lets you:

• See health of each project at a glance
• Filter the dashboard to only see one project's monitors
• Create a dedicated status page per project
• Set alert rules scoped to a specific project
• Share one monitor (e.g. Stripe) across multiple projects

Creating an Application

1. Go to Dashboard → click the Application switcher at the top
2. Select "Create Application"
3. Choose a name, icon and color to identify your project
4. Use the wizard to add catalog APIs, custom endpoints, websites and heartbeats in one step
5. Your new Application appears in the switcher immediately

Plan Limits

Website Monitors

Website Monitors check any URL for uptime, response time, SSL certificate validity and HTTP status — independently of your API monitors. Ideal for landing pages, marketing sites and web apps.

How to Add a Website Monitor

1. Go to Dashboard → Websites → Add Website
2. Enter the URL (e.g. https://yourcompany.com)
3. Optionally assign to an Application
4. Save — monitoring starts immediately

What's Monitored

• Uptime: Checked from multiple regions every 5 minutes
• Response time: Average latency with historical chart
• SSL certificate: Expiry date tracking with advance alerts (30, 14, 7 days)
• HTTP status: Any non-2xx response creates an incident

Plan Limits

Stack Pulse

Stack Pulse connects directly to your infrastructure providers via their APIs to surface real operational metrics — not just "is this endpoint up?" but "how many deployments are active?", "is the database healthy?", "how many emails were delivered in the last hour?". Included on every plan — the free plan carries up to 3 integrations.

Supported Providers

How to Add an Integration

1. Go to Dashboard → Stack Pulse
2. Click "Add Integration" and choose your provider
3. Enter the required API token or key (read-only credentials are recommended)
4. Probecast verifies the credentials and starts collecting immediately

Plan Limits

Revenue & Growth (paid plans)

Connect your payment provider and Probecast shows your real revenue right next to your uptime — per product. We pull live data from your gateway so you can watch MRR, today's revenue and growth in the same place you watch reliability and tie incidents to their revenue impact.

Supported Gateways

Each gateway has its own capabilities — recurring revenue (MRR) support, how it connects, and whether one account can be split across products. Here's what to expect from each:

Running client sites? Shopify and WooCommerce stores connect read-only per client — see the Store Revenue section under Client Work below.

Where You'll See It

• Main View — each product shows today's revenue alongside its health status.
• Wallboard — per-product revenue tiles on the big-screen ops view.
• Revenue page — MRR, revenue trends and growth, broken down by product.
• Incident Correlation (Agency) — the revenue impact of downtime, linking incidents to the revenue at risk.

How to Connect

1. Go to Dashboard → Revenue and choose your provider
2. Authorize with your gateway or paste a read-only API key
3. Map the connection to the product it belongs to
4. Probecast verifies the credentials and starts pulling revenue immediately

Revenue & Growth analytics are included on all paid plans. AI Revenue Insights are available on Startup and above and Revenue Correlation with SLA targets is part of Agency.

Probecast Pixel

Probecast's own first-party tracking script — the alternative to Google Analytics for the Growth section. One small script per product: live visitors, daily traffic, channels, campaigns (UTM), funnel and journey tracking, tied to your real revenue. Cookieless by default: until you ask for consent, it uses a rotating session identifier — no persistent ID, no PII, no third parties.

Consent (optional)

If you run a cookie banner, call probecast.consent('granted') on acceptance to promote the visitor to a first-party cookie (returning visits get recognised), and probecast.consent('denied') on rejection — that wipes the persistent ID and returns to the rotating one. Declined means declined: there is no respawn.

Opting out of tracking

Any visitor — including you, so your own visits don't pollute your numbers — can turn the pixel off entirely for their browser:

The opt-out is stored in both a cookie and localStorage, applies regardless of consent state, and is resolved before any identifier is created or any event is sent — an opted-out browser sends nothing at all.

Health Score per Application (paid plans)

Each Application gets a Health Score (0–100) — a single number that represents the overall reliability of that project, recalculated every 5 minutes.

How It's Calculated

• Uptime (40%): Average uptime of all monitors in the Application over the last 24h
• Latency (25%): Inverse of average latency — lower latency = higher score
• Incidents (20%): 0 active incidents = 100; each incident reduces the score
• Coverage (15%): % of monitors with a recent check (last 15 minutes)

Interpreting Your Score

• 80–100: Healthy — all systems normal
• 50–79: Degraded — some services impacted
• 0–49: Critical — action needed

Public Badges (paid plans)

Showcase your API reliability with embeddable badges for your website, README or documentation.

How to Use

Each monitored API has a unique badge you can embed:

Customization Options

• Period: 24h, 7 days, 30 days or 90 days
• Style: Badge, Shield or Flat
• Color: Automatic based on uptime or custom
• Label: Customize the displayed text

Example Code

Badges update automatically in real-time. When your API goes offline, the badge reflects this immediately.

Public Status Pages

Create public status pages for your customers to monitor the health of your services in real-time.

What Are Status Pages?

Status Pages are public web pages (like status.yourcompany.com) that show the current and historical status of your services. Your customers can:

• View real-time status of all your services
• Check incident history and uptimes
• Receive notifications when there are problems
• Access 24/7 without requiring login

How to Create a Status Page

1. Go to Dashboard → Public Status Pages
2. Click "Create New Status Page"
3. Select which APIs you want to display publicly
4. Customize: title, description, logo, colors
5. Your page is published at probecast.io/public-status/your-slug
6. Publish!

Custom Domain (Agency)

Agency plan customers can use their own domain:

• status.yourcompany.com (instead of yourcompany.probecast.io)
• Configure CNAME in your DNS pointing to Probecast
• Automatic SSL/HTTPS via Let's Encrypt

Heartbeat Monitoring

Heartbeat Monitoring is an inverse monitoring approach: instead of Probecast checking your service, your service sends periodic pings to Probecast. If a ping is missed, an alert is triggered.

How It Works

Ideal for monitoring cron jobs, background workers, batch processes, scheduled tasks and any service that runs on a schedule.

Integration Examples

# Cron job example
0 * * * * /path/to/your/script.sh && curl -s https://probecast.io/api/heartbeat/ping/your-slug

# Node.js example
await doWork();
await fetch('https://probecast.io/api/heartbeat/ping/your-slug');

# Python example
do_work()
requests.get('https://probecast.io/api/heartbeat/ping/your-slug')

Limits per Plan

Webhooks (Startup & Agency)

Integrate Probecast with any system through HTTP webhooks. Receive real-time notifications of incidents.

Webhooks push alerts out; to pull data or create monitors from code, see the REST API, MCP server and CLI.

How They Work

When an incident occurs (or is resolved), Probecast makes an HTTP POST to the URL you configure, with all event data.

Configuration

1. Go to Settings → Webhooks
2. Click "Add Webhook"
3. Enter your endpoint URL
4. Choose which events to receive (incident.created, incident.resolved, etc)
5. Configure optional secret to validate signatures
6. Save and test

Example Payload

{
  "event": "down",
  "severity": "critical",
  "source": "website",
  "api": { "id": "mon_abc123", "name": "fashionstore.com" },
  "message": "fashionstore.com is DOWN (HTTP 503)",
  "details": { "region": "europe", "status_code": 503 },
  "timestamp": "2026-08-14T18:30:15Z"
}

Need a different shape? Alert-rule webhook channels accept an optional body template — see the webhook recipes below.

Popular Integrations

• Slack: Send messages to specific channels
• Custom: Any HTTP endpoint that accepts POST

Webhook Recipes — incident → ticket

A confirmed incident can open a ticket or task in the tool your team already lives in. These recipes use the alert-rule webhook channel (Settings → Channels → Webhook) with custom headers and an optional body template — no middleware, no extra services.

Custom body template (any ticket or task tool)

If your tool accepts ticket or task creation over HTTP, point a webhook channel at its endpoint, add its auth header and shape the JSON body with a template:

{
  "title": "[{{severity}}] {{source_name}}: {{message}}",
  "created_at": "{{timestamp}}",
  "probecast": {{details_json}}
}

Variables are JSON-escaped automatically — quotes and line breaks in the incident message never break the body. Available: {{event}}, {{severity}}, {{source_name}}, {{message}}, {{timestamp}} and {{details_json}} (the raw details object).

HelpScout (email-to-ticket)

HelpScout creates conversations from inbound email natively — that is the most reliable path (its API requires short-lived OAuth tokens that a static webhook cannot refresh). Add an email channelpointing at your mailbox's forwarding address (HelpScout → Mailbox → Settings → Forwarding address) and use a custom alert template on the rule so the subject reads like a ticket title — e.g. [{{severity}}] {{api_name}} {{status}}. Tag-based workflows in HelpScout can then route it to the right team.

Gorgias (email-to-ticket)

Same pattern: Gorgias turns inbound email into tickets natively. Point an email channel at your Gorgias intake address and let a Gorgias rule tag incidents (e.g. incident) so they skip the support queue metrics. Pairs naturally with the Shopify revenue integration for e-commerce clients.

Incident Reports

Each incident generates a complete report with detailed analysis and recommendations.

What's Included

How to Access

Reports are available in:

• Dashboard → Incidents → Click on any incident
• Via direct link in email/Slack notifications

Alert Grouping

When multiple APIs from the same vendor fail simultaneously, Probecast consolidates them into a single notification instead of sending separate alerts for each API.

How it works

• Alerts are buffered for 6 minutes after the first failure
• If multiple APIs from the same vendor fail in that window, they are grouped into one notification
• Recovery alerts are always sent immediately (not buffered)
• If a recovery arrives during the buffer window, that API is removed from the group

Example: If Stripe Payments, Stripe Checkout and Stripe Invoices all fail within 6 minutes, you receive one email listing all three — not three separate alerts.

Available on all plans.

Custom Alert Templates

Customize the subject line and body of email alerts sent by your alert rules. Useful for integrating with ticketing systems or matching your team's communication style.

Available variables

Configure templates in Settings > Alert Rules when creating or editing a rule. Available on all paid plans.

Maintenance Windows

Schedule planned maintenance periods where health checks continue running but alert notifications are silenced. This prevents false alerts during deployments, migrations or infrastructure changes.

Key features

• Monitoring continues: Unlike some competitors that stop monitoring entirely, Probecast keeps collecting health data during maintenance
• Recurrence: Schedule one-time, daily, weekly or monthly windows
• Scoped: Apply to all monitors, specific APIs or a specific product
• Overnight support: Windows that span midnight (e.g., 23:00–02:00) work correctly

Configure in Settings > Maintenance Windows. Available on all paid plans.

Auto-Incident Posts

Automatically create incident posts on your public status page when monitors detect downtime. No manual intervention required — your status page stays up to date in real time.

How it works

• Enable the toggle in your Status Page settings
• When an incident is created, a post is automatically added to your status page
• When the incident resolves, a resolution post is auto-created
• You receive an email notification with Edit and Delete buttons for each auto-post
• Auto-generated posts are marked with a badge so you can distinguish them from manual posts

Available on all paid plans.

White-Label & Custom Domain

Remove Probecast branding from your status pages and alert emails on any paid plan. On the Agency plan, you can also host your status page on your own custom domain.

Paid plans (Indie, Startup & Agency)

• Remove "Powered by Probecast" from public status pages
• Remove [Probecast] prefix from alert email subjects
• Remove Probecast branding from email headers

Agency plan — Custom domain

Host your status page on your own domain (e.g., status.yourcompany.com).

DNS setup instructions

Smart Alerts (AI)

Probecast uses AI-powered analysis to reduce alert noise. Before creating an incident, the system evaluates whether the failure is a real outage or a transient blip.

How it works

• Severity scoring (1-10): Based on status code, consecutive failures, affected regions, uptime history and time of day
• False positive detection: Identifies single-check blips on high-uptime APIs, single-region failures and rate limiting (429)
• AI summary: Optional Claude-powered analysis providing a concise explanation of the alert with suggested actions

When alerts are suppressed

• Single failure on a 99.9%+ uptime API (likely transient)
• Only one region affected on a normally stable API
• HTTP 429 responses (rate limiting, not a real outage)

Smart Alerts never block a legitimate incident — if the AI analysis times out or fails, the incident is created normally. A 2.5-second circuit breaker ensures health check processing is not delayed.

Available on all plans.

Synthetic Monitoring (Agency)

Synthetic checks run a real browser against your site on a schedule and assert that a flow actually works — not just that the page answers. Describe the flow in plain English and Probecast compiles it into a Playwright script for you.

How it works

• Write the flow as prose — e.g. “open pricing, click Start trial, confirm the checkout loads”
• Probecast generates the script, shows you every step and runs it from the monitoring worker
• A failed run captures a screenshot of the exact moment, attached to the incident
• Failures are confirmed on a second run before alerting, like every other monitor

The sandbox runs navigation, clicks and visibility assertions. Keyboard input, hover, scrolling and persistent sessions are not supported — keep flows to what a first-time visitor could do. Agency includes 10 synthetic monitors.

Monthly Client Reports (Agency)

One document per client per month: availability, incidents, traffic, revenue, support and the third-party dependencies underneath — drafted by Probecast, reviewed and signed off by your team, delivered under your brand.

The flow

• Generate: Dashboard → Client reports → pick the client and the month. The current month works too, as a clearly-marked preview.
• Review: every report starts as a draft. Drafts never leave the workspace — no link, no email, until someone finalizes.
• Finalize: one click when the month reads right. A final report is frozen — data, narrative and sections stop changing.
• Deliver: share a private link (revocable, token-based) or email it straight to the client's contacts. Print to PDF from the browser.

What's in the document

• Day-by-day availability, incidents with duration and severity, total downtime
• Third-party dependencies with logo and uptime for the period
• Traffic (pixel), revenue (store or gateway) and support — tickets opened, resolved, still open at month end, volume vs the last 3 months, optional ticket-id annex
• “The month in words” — a written narrative in the report's language, plus your account manager's note

Make it yours

• Six languages: English, Spanish, French, German, Dutch and Brazilian Portuguese
• Three templates (Executive, Detailed, Minimal) + per-section toggles
• Your agency logo and “prepared by” name; the client's logo in the header
• The “Independently monitored by Probecast.io” line is on by default — switch it off per client in the report settings

Helpdesk Integrations (Agency)

Connect HelpScout, Zendesk, Gorgias or Freshdesk read-only and map each client by mailbox or tag. Support work becomes part of the client's monthly report — and an urgent ticket becomes an alert in real time (HelpScout and Gorgias).

What Probecast reads — and what it never reads

• Counts and timings only: tickets by severity and type, first-response time, resolution counts
• Never a subject line, a message body or a customer's name — content stays in your helpdesk
• The optional report annex lists ticket identifiers only, so your client can pull any case by its number

Setup: Dashboard → the client's space → Support → connect the provider with a read-only token, then map the mailbox or tag that belongs to that client.

Store Revenue — Shopify & WooCommerce

Connect a client's store and their orders and revenue appear next to uptime — in the dashboard and in the monthly report. Built for agencies that run stores they don't own.

• Shopify: connect with a read-only scope — Probecast verifies the scope on connect and refuses credentials that could write
• WooCommerce: connect with a REST key — a write probe runs on connect and read-write keys are rejected
• Revenue is grouped per currency and attributed to the client the store belongs to

The read-only rule is enforced by the product, not by policy: a credential that could modify the store never gets saved.

Team & Audit Log (Agency)

Invite your team and scope each person to the clients they work on. Seats never change the price.

Scoping

• Per client: a freelancer assigned to two clients sees those two clients — nothing else
• Revenue is a separate permission: a member can manage monitors without ever seeing money numbers
• Read-only: give auditors or clients' stakeholders a view that cannot change anything

Audit log

Every change lands in an append-only audit log — enforced by the database itself, so entries cannot be edited or deleted after the fact. Filter by action, client or person and export to CSV from Dashboard → Audit.

REST API

Everything the dashboard does for monitors, scripted. Create an API key in Settings → API Keys (paid plans) and send it as a Bearer token. Read keys list monitors and incidents; write keys can also create and delete monitors. Full contract: openapi.yaml.

# Who am I and what fits in my plan?
curl https://api.probecast.io/v1/me \
  -H "Authorization: Bearer pb_live_..."

# Add a website monitor for a client (creates the client if missing)
curl -X POST https://api.probecast.io/v1/monitors \
  -H "Authorization: Bearer pb_live_..." \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: deploy-acme-2026-10-25" \
  -d '{"type":"website","name":"acme.com","url":"https://acme.com","client":"Acme"}'

# Anything on fire?
curl https://api.probecast.io/v1/incidents \
  -H "Authorization: Bearer pb_live_..."

• Idempotency: send an Idempotency-Key on POST /monitors — a retried request returns the original response instead of a duplicate
• Quotas: the API enforces the same plan limits as the dashboard; GET /v1/me shows what is left
• Rate limits: 120 reads and 30 writes per minute per key
• Heartbeats: creation returns a ping_url — call it from the cron itself, no key needed there

MCP Server

Let your coding agent set up monitoring while it ships the project. The server runs locally via npx, talks to the API with your key and exposes six tools: add_monitor, list_monitors, remove_monitor, get_incidents, get_status and get_quota. Zero dependencies — the whole server is two small files you can read.

# Claude Code
claude mcp add probecast -e PROBECAST_API_KEY=pb_live_... -- npx -y @probecast/mcp

# Cursor / Windsurf (mcp.json)
{ "mcpServers": { "probecast": {
  "command": "npx", "args": ["-y", "@probecast/mcp"],
  "env": { "PROBECAST_API_KEY": "pb_live_..." }
} } }

Then, in the editor: “add a Probecast monitor for acme.com under the client Acme” — the agent calls add_monitor and the monitor appears in your dashboard. We also monitor public MCP servers themselves: see the MCP status hub.

CLI

The same API from the terminal — handy in deploy scripts and CI.

npx probecast login pb_live_...   # stored in ~/.probecast (0600)
npx probecast add --type website --name "acme.com" --url https://acme.com --client "Acme"
npx probecast add --type heartbeat --name "nightly backup" --interval 60
npx probecast incidents
npx probecast status github-mcp

Billing & Plans

Available Plans

Four plans, two audiences: Free, Indie ($9) and Startup ($25) for founders; Agency ($150, 40-day trial) for agencies and studios — the first agencies are onboarded personally. Full details at probecast.io/pricing.

Changing Plans

You can upgrade or downgrade at any time in Settings → Billing. Changes are applied with automatic proration - you only pay the proportional difference for the remaining time in the cycle.

Payment Methods

We accept all major credit cards through Stripe. All transactions are secure, encrypted and processed in compliance with PCI-DSS.

Failed Payments

If a payment fails, we don't cut you off right away. Stripe automatically retries the charge over the following days and we email you at each step so nothing catches you by surprise:

• An initial notice the moment a charge fails
• Reminders while automatic retries are still pending
• A final warning before the subscription is canceled

You can resolve it at any point by updating your card in Settings → Billing. If every retry fails, the subscription is canceled and your account enters the 60-day retention window described below — your plan tier is never silently downgraded and your data is kept intact.

Cancellation

You can cancel at any time, without penalties or notice periods. Your plan stays active until the end of the billing period you've already paid for (a canceled subscription is not refunded for the remaining time). After that, your account enters a 60-day retention window: your products, monitors, history and settings are preserved and you can reactivate with one click to pick up exactly where you left off. If you don't reactivate within 60 days, the account and its data are permanently removed.

Account Suspension

In rare cases — such as a payment dispute (chargeback) or a refund — we may temporarily suspend access to your account. When this happens your plan, data and configuration stay fully intact; nothing is deleted or downgraded. Access is restored automatically once the issue is resolved (for example, when a dispute is closed in our favor). We'll always email you if your access is suspended, with the reason and how to fix it.

Need Help?

Contact our support team: support@probecast.io

Data & Privacy

We take your privacy seriously. We are compliant with GDPR and data security best practices.

GDPR Compliance

Probecast is fully compliant with the General Data Protection Regulation (GDPR). You have the right to:

• Access: View all data we store about you
• Rectification: Correct incorrect or outdated data
• Portability: Export your data in JSON format
• Deletion: Request complete removal of all your data

Account Deletion

You can request deletion of your account at any time in Settings → Account → Delete Account.

After deletion confirmation:

• All personal data is permanently removed
• Monitored APIs, history and settings are deleted
• Paid subscriptions are canceled (no prorated refund)
• Complete process within 48 hours

Data Security

• Encryption: TLS 1.3 in transit, AES-256 at rest
• Headers and Credentials: Never logged, always encrypted
• Backups: Daily encrypted backups, retained for 30 days
• Access: Minimum privilege, 2FA mandatory for admins
• Audits: Audit logs for all sensitive operations

Data Retention

Frequently Asked Questions (FAQ)

Answers to the most common questions about Probecast.