Skip to main content
JUMP TO:Quick StartCLI CommandsConfigurationPlaywright APIMCP ServerGitHub Actions
Guides / Production Monitoring(17 of 37)

Production Monitoring

LOCAL ONLY. The implemented path is frontguard monitor run by your own process, cron, or CI. A managed cloud scheduler is not generally available.

Use frontguard monitor to run synthetic visual checks against live production or staging URLs, record timestamped run history on disk, and dispatch webhook alerts when visual drift exceeds your threshold.

Frontguard Monitor Production Monitoring and Alert Pipeline
FIG 01 · PRODUCTION MONITORING PIPELINE — CRON/DAEMON TRIGGER · HERMETIC CAPTURE · FLAKE TRACKER · HISTORY & WEBHOOKS

Local CLI

terminal
$ frontguard monitor --url https://example.com --threshold 0.05 --once
$ frontguard monitor --history
FlagDefaultDescription
--url <url>baseUrl from configLive production or staging URL to monitor.
--threshold <ratio>0.05 (5%)Changed-pixel ratio required to trigger a monitor alert (0-1 scale).
--oncefalseExecute a single check and exit immediately (ideal for Kubernetes CronJobs or GitHub Actions schedule).
--interval <minutes>unsetPoll every N minutes. Without --interval or --watch, run once.
--watchfalseContinuously poll; default interval is one minute unless --interval is provided.
--webhook <url>undefinedHTTP endpoint accepting Frontguard JSON alerts. Slack Incoming Webhooks require an adapter; see the Slack integration guide.
--historyfalsePrint the local run history log stored under .frontguard/monitor-history.

Scheduled cron recipe

Because frontguard monitor --once exits non-zero when a production regression exceeds threshold, you can run it on a 15-minute or hourly schedule in GitHub Actions or any cron runner:

.github/workflows/prod-monitor.yml
name: Production Visual Monitor
on:
  schedule:
    - cron: '0 * * * *' # Every hour
  workflow_dispatch:

jobs:
  monitor:
    runs-on: ubuntu-latest
    permissions:
      contents: read
    steps:
      - uses: actions/checkout@v4
      - uses: actions/setup-node@v4
        with:
          node-version: 22
      - run: git fetch origin frontguard-baselines:frontguard-baselines
      - run: npx -p @frontguard/cli playwright install --with-deps chromium
      - run: npx -p @frontguard/cli frontguard monitor --url https://example.com --threshold 0.05 --once

History

CLI monitor records run history under .frontguard/monitor-history by default. Webhook failures are reported but should not be treated as a durable paging system.

Hosted status

Self-triggered cron or CI execution is the supported production monitoring workflow today.