Dashboard guide
Learn how the PreFlight dashboard is organized, what each feature tab does, and how to debug checks.
Guides
Command Center
The PreFlight dashboard is organized around the questions a launch team asks before and after a deploy: Is the workspace healthy? Which provider needs attention? What failed, and should we block the release? Can we show proof to a customer?
Every surface reflects stored probe rows, samples, and audit events — never demo telemetry. Empty sections stay empty until your workspace has real projects, integrations, checks, alerts, or reports.
How navigation is structured
The sidebar groups features by job, not by provider. Overview and Projects anchor the workspace. Watchdogs cover post-deploy sampling. Operations holds release controls, runbooks, and evidence. Configuration is where you connect providers, export reports, and manage keys.
Use Cmd/Ctrl+K to jump to a project, incident, or alert without hunting through tabs.
Workspace surfaces
Overview
Overview is the starting point. It shows the workspace health banner, setup progress, readiness score, provider coverage, seven-day trend, open issues, and recent activity.
Use Overview when you need the shortest answer to "are we okay right now?"
Projects
Projects are the apps PreFlight is allowed to check. Each project owns its saved URL, connected providers, check history, monitoring sessions, guardrails, Trust Center, and settings.
Project tabs map directly to launch workflows:
| Tab | Route | Primary question |
|---|---|---|
| Overview | /dashboard/projects/:id | What is this project's current posture? |
| Checks | …/checks | What did the latest run prove? |
| Integrations | …/integrations | Which providers are connected and when were they last tested? |
| Monitoring | …/monitoring | Is the 24-hour release window active? |
| Guardrails | …/guardrails | Are schema sync and release guardrails clean? |
| Trust Center | …/trust-center | What can we publish to customers? |
| Settings | …/settings | URL, environment labels, and project metadata |
The readiness URL scanner only uses the URL saved on the project. Do not point it at unrelated domains.
Pre-Flight Checks
Pre-Flight Checks (/dashboard/checks) run one deterministic suite across your saved app URL and every connected provider. Each probe returns success, warning, or failed with response time and a fix suggestion.
Checks are split into focused views:
| View | What it covers |
|---|---|
| Full runs | The complete standard check for a project |
| App URL checks | DNS, SSL, SEO, legal pages, performance, and Exposed Secrets scanning for leaked keys in HTML/JS bundles |
| Provider checks | Credentials, callback URLs, permissions, quotas, and provider-specific reachability — including Webhook diagnostics for Stripe, Lemon Squeezy, and Discord |
| Fix list | Failed and warning probes only |
| AI Review | Optional Cerebras-backed diagnosis built from stored probe rows, with deterministic fallback if AI is unavailable |
Disconnected providers appear as skipped so the run is easy to explain to the team.
Test Checkout
Shadow mode validates the full checkout loop without a real card: PreFlight creates a test-mode Checkout session, verifies webhook delivery, and confirms the Supabase side effect landed. Configure it on the Stripe integration and run it from Checks or CI. See Verify Stripe Checkout without a card.
Watchdogs
Watchdogs sample health after a check passes. Use them in order: full check → 24-hour Monitoring → Sentinel.
24-hour Monitoring
24-hour Monitoring (/dashboard/monitoring) is a time-boxed release-window watcher. Start it before a launch or immediately after a risky deploy. It samples the health endpoint and connected providers with denser cadence, routes alerts on repeated failures, and completes automatically after 24 hours.
Uptime Monitor
Uptime Monitor (/dashboard/sentinel) is always-on production sampling. Each project can be enabled, paused, assigned a cadence, and given a failure threshold. It catches drift after the launch window closes.
Vercel Watch
Vercel Watch (/dashboard/vercel-watch) continuously verifies connected Vercel projects: production deployment status, domain drift, and environment parity before users notice.
Revenue Watch
Revenue Watch (/dashboard/revenue-watch) reconciles paid Stripe Checkout sessions against Supabase side effects configured for shadow checks. It surfaces silent revenue drift when webhooks fail or write to the wrong table.
See Revenue Watch.
Data Integrity Guard
Data Integrity Guard (/dashboard/backups) captures a metadata-only baseline of your Supabase database, detects paused projects, dropped tables, and collapsed row counts, then runs a read-only integrity drill. It does not claim that a provider backup is restorable.
Readiness and Custom Checks
Readiness (/dashboard/readiness) focuses on the saved app URL and launch probes in one view. Custom Checks (/dashboard/custom-checks) let you schedule your own HTTP assertions on the monitoring cron cadence.
Operations
Incidents
Incidents (/dashboard/incident-command) groups failed uptime samples, provider probes, alert deliveries, and runbook state into actionable work items. Use it when the question shifts from "what failed?" to "what are we actively handling?"
Runbooks
Runbooks (/dashboard/runbooks) map failed and warning probes to rule-based recovery steps with verification instructions and docs links. Each row explains why the probe failed, where to check, and how to confirm recovery.
Deploy Gates
Deploy Gates (/dashboard/deploy-gates) expose hard pass/fail release checks for CI and Vercel. Gates can require clean checks, active Sentinel, clean Revenue Watch, schema sync, and provider health.
See Deploy Gates.
Provider SLA
Provider SLA (/dashboard/provider-sla) summarizes seven-day reliability per integration: pass rate, warning rate, failure count, average latency, and most recent check time.
Incident Flight Recorder
History & Audits (/dashboard/activity) is the Flight Recorder timeline: checks, deploys, provider events, revenue snapshots, alerts, and settings changes correlated into a replayable audit trail.
API Keys & Security
API Keys & Security (/dashboard/api-security) inventories active API keys, rejected boundary events, forged Shield reports, and durable rate-limit buckets.
Supabase schema sync
Schema sync lives under Guardrails on each project and in the global guardrails view. It detects when a Supabase migration diverges from the deployed app and can block releases through Deploy Gates.
Alerts and evidence
Alert Routing
Alerts (/dashboard/alerts) shows delivery history for Slack, Discord, email, PagerDuty, OpsGenie, Microsoft Teams, and generic webhooks. Confirm whether a message was sent, failed, bounced, or skipped by notification preferences.
Reports and exports
Exports (/dashboard/exports) produces PDF, CSV, JSON, and expiring share links backed by stored check runs — auditor-friendly evidence without credentials or raw webhook payloads.
Trust Center
Trust Center is project-scoped customer-facing evidence at /dashboard/projects/:id/trust-center. Publish health summaries and uptime without exposing provider secrets.
See Trust Center.
Configuration
Providers
Providers (/dashboard/providers) is where you connect Stripe, Supabase, Clerk, Vercel, Resend, and the rest of the integration catalog. Each connection stores encrypted credentials and drives the provider probe suite.
Account API + GitHub guardrails
Settings → API keys creates account-scoped keys (pf_live_…) for CI automation: trigger checks, read gate state, and ship reports from your pipeline. See API overview.
Settings
Settings manages workspace team invites, alert channels, API keys, on-call routing, password reset, and restricted destructive actions.
Recommended audit flow
When debugging a launch issue, follow this sequence:
Open Fix list. In Checks, filter to failed and warning probes only.
Clear failures first. Failed probes block strict Deploy Gates and should open a runbook.
Read the probe message. Each row includes response time, safe error code, and fix suggestion.
Apply the fix and rerun. Run the full check again — partial reruns are available for individual providers.
Check downstream surfaces. If the issue affects release automation, review Sentinel, Revenue Watch, Vercel Watch, and Deploy Gates.
Export or publish. Generate an export or Trust Center report once the result is clean enough to share.
App URL vs provider checks
Use App URL checks for DNS, SSL, legal pages, and secret leaks in public bundles. Use Provider checks for credentials, webhooks, callbacks, and API permissions.
Product surface reference
Every shipping module in PreFlight maps to a dashboard route:
| Module | Dashboard | Tagline |
|---|---|---|
| Pre-Flight Checks | Checks | One run, every fragile path validated |
| 24/7 Sentinel | Sentinel | Always-on health sampling |
| 24-hour Monitoring | Monitoring | Dense sampling during launch week |
| Incident Command | Incidents | Failures grouped into work items |
| Vercel Watch | Vercel Watch | Deployments and domains, continuously verified |
| Revenue Ledger Watch | Revenue Watch | Stripe to Supabase, reconciled |
| Stripe Shadow Check | Checks + Stripe integration | Full checkout loop without a real card |
| Data Integrity Guard | Data integrity | Independent metadata baseline and read-only integrity drill |
| Deploy Gates | Deploy Gates | CI and Vercel release blocking |
| API Security Center | API Security | Keys, boundaries, and rate limits |
| Secret Drift Scanner | App URL checks | Find leaked keys before launch traffic |
| Provider SLA | Provider SLA | Seven-day pass rate per integration |
| Incident Flight Recorder | History & Audits | Persisted timeline, every release |
| Deterministic Runbooks | Runbooks | Fix guidance that stays inside facts |
| AI Launch Review | Checks → AI diagnosis | Optional model assist on probe data |
| Alert Routing | Alerts | Slack, Discord, and email channels |
| Trust Center | Project → Trust Center | Customer-facing health evidence |
| Reports and exports | Exports | PDF, CSV, JSON, and share links |
| Webhook diagnostics | Provider checks | Signature/handshake verification and delivery checks |
| Supabase schema sync | Guardrails | Catch migration drift before it ships |
| Account API | Settings → API keys | Stable endpoints for CI automation |