Testing & Scheduling
Two things turn a configured source into a reliable, hands-off part of your estate: testing that it connects, and scheduling it to scan on its own.
Test the connection
Before committing to a full scan, you can test the connection to check that the details you entered actually work — without reading any content or producing findings. It’s the fast way to catch a typo’d host, an expired token, or a permission problem.
A test returns one of two clear outcomes:
| Result | Meaning |
|---|---|
| Success | Classifyre reached the system and authenticated. You’ll get a short confirmation message (e.g. “Successfully connected to Slack workspace acme.”). |
| Failure | Something’s wrong, with a human-friendly message pointing at the likely cause so you can fix it. |
Why it’s especially useful here: because secrets are masked (write-only and never shown back), testing is the proper way to confirm a credential is correct — you verify it works rather than trying to re-read it.
Test right after creating a source, and again whenever you rotate a credential or change connection details. A quick test now saves a failed scan later.
Scan on a schedule
A source can scan on a recurring schedule so coverage stays current with no one having to remember to press a button.
Scheduling has three parts:
| Setting | What it does |
|---|---|
| Enabled | Turns the recurring schedule on or off. |
| Cadence | When to run, as a standard cron expression (e.g. every night, every weekday morning). |
| Timezone | The timezone the cadence is interpreted in (defaults to UTC). |
You can still run a scan manually at any time — the schedule simply adds automatic runs on top.
Scheduling + Automatic sampling = effortless coverage
Recurring scans are most powerful paired with the Automatic sampling strategy:
Each scheduled run reads a bounded slice, picks up where the last one stopped, and — over several runs — covers the entire source, then loops back to catch what changed. A little, often, automatically.
What happens after a scan
Testing and scheduling get a source running; the rest of the platform takes it from there:
- Flow — what a scan does step by step, and how repeat scans diff against the previous run.
- Investigations — how findings become inquiries and cases.
- Autopilot — the AI agents that act on new findings after each scan, and can even tune a source’s detectors for you.
Keep going
| Page | |
|---|---|
| How Sources Work | What a source is and its journey to findings |
| Configuration & Fields | Required, masked, and optional fields |
| Sampling Strategies | How much to read, and which items |
| OCR & Transcription | Reading images, audio, and video |
| Testing & Scheduling | Verify connections and automate scans (you are here) |
| Assets & Metadata | What a scan produces |
See what a scan actually yields in Assets & Metadata, or look up a specific system’s exact fields in the Source Catalog.