Content filter

Automatically check what people write in your Hub for harmful content, and review what it flags.

What the content filter does#

Truck Pulse checks free text people write — your Hub's description, event descriptions, application answers, reports, staff messages — for harmful content before it reaches anyone. It runs automatically on every Hub. There is nothing to set up.

Two things can happen:

• Clearly over the line — the text is refused as it is typed. Whoever wrote it sees a short message saying what the problem is, and can reword and try again. Nothing is saved and nobody is notified.
• Borderline — the text saves normally, and a copy lands in Members → Flagged content for your staff to look at.

Where you draw the line between those two is up to you — see Settings below.

What it does not catch: ordinary swearing. The filter looks for hateful, harassing, sexual, violent and self-harm content, and for talk about illegal activity. Somebody swearing in an event description will not trip it. If you want a stricter tone than that, say so in your Hub rules and handle it the way you handle any other rule.

Reviewing what it flags#

Flagged text appears in Members → Flagged content, next to driver reports, with a badge counting what is still waiting. Each row shows when it was flagged, which record and field it came from, an excerpt, what tripped the filter, and how confident it was.

Open one to see the whole picture, including one particularly useful line: Has it changed since? A "Yes" means the author has already edited the text, so there is often nothing left to do.

From there:

1. Take under review — mark that you are looking at it.
2. Uphold — the filter was right. Close the flag with a note saying what you did about it.
3. False alarm — close it as fine. Do use this rather than leaving it: clearing a flag also stops that same text being flagged again.
4. Escalate to the platform — send it to Truck Pulse staff as an abuse report, for content that goes beyond something your own Hub should be handling.

Editing the text itself happens on the record's own page, exactly as it normally would. A flag is a pointer to a problem, not a place to fix it.

Every decision is recorded in your audit log with your name against it.

Who can see flagged content#

Two permissions control this, granted through roles:

Owners hold both. Nobody else does until you grant them — including the person whose text was flagged, who never sees their own flag.

Settings#

Settings → Modules → Content filter.

Check what people write for harmful content turns the whole thing on or off for your Hub.

Below it are five kinds of content, each with its own setting:

Hate & harassment, Sexual content and Self-harm start at Standard. Violence and Illegal & criminal activity start at Lenient, and deliberately so: truck-sim Hubs talk about crashes, pile-ups, wrecked trailers and reckless driving reports all day, and stricter settings here mostly produce work for your staff rather than safer text.

Whatever you choose, sexual content involving minors is always refused, in every Hub. That one is not configurable.

Tell staff when something is flagged sends a notification, and a Discord message if you have notifications set up, whenever something lands in the queue.

Under Advanced, Only refuse text on public pages changes what happens to text your staff write: instead of being refused, it saves and goes to the queue. Text from drivers and visitors is still refused either way. Turn this on if your staff find the filter getting in the way of writing event copy.

When the checker is unavailable#

The filter depends on an outside service. If that service is slow or down, text saves normally rather than being blocked — nobody is ever stopped from saving their work by somebody else's outage. Anything that could not be checked is recorded in your queue as Not checked and is checked again automatically shortly afterwards.