Censorship and abuse warnings
Mask configured abusive phrases, warn visitors and block repeat abuse according to website policy.
What it is
Censorship replaces matched abusive phrases with asterisks and can warn the visitor. Repeated matched messages can block the visitor, using the threshold set for your website.
Who uses it and why
Powerfront sets the phrase lists, warning text and blocking policy for you. Operators and supervisors see filtered messages and the resulting warnings or blocked state. Visitors receive warnings when configured.
Where to find it
Powerfront manages Settings › General Settings › Censorship and can change the configuration for you.
How it works
INSIDE checks visitor messages against the system default phrases and your additional Blocked Phrases, allowing exceptions in Whitelist Phrases. Matching ignores letter case and uses word boundaries. A matched phrase is replaced by the same number of asterisks.
With the default threshold of three matched messages, the first receives System Warning Message, the second receives Final Warning Message, and the third triggers blocking. Empty warning text does not provide a visible warning.
Web chat shows the visitor a warning in the chat experience. Other messaging platforms use an automatic reply, while email uses the ticket email flow.
If the visitor belongs to a linked contact, automatic censorship blocking applies to that contact's visitors. For an unlinked visitor, blocking applies to that visitor.
Configure it
Ask Powerfront to enable Enable Censorship and set warning text and the blocking threshold. Powerfront maintains warning text and phrase lists on each language tab and can inspect the default phrases with Show System Default Blocklist.
Ask Powerfront to set Block for (x) days according to your policy. Zero means indefinite blocking. A positive duration lets INSIDE clear the censorship block after the period has elapsed, including the linked contact block where applicable. Powerfront saves the changes with Save.
Settings
| Setting | Where | Applies to | Default | Who can change it | What it does |
|---|---|---|---|---|---|
| Enable Censorship | Settings › General Settings › Censorship | website | Off | Powerfront | Filters visitor messages against default and custom blocked phrases. |
| System Warning Message | Settings › General Settings › Censorship | website | Not set | Powerfront | Sets the warning sent before the final warning stage, with language-specific text. |
| Final Warning Message | Settings › General Settings › Censorship | website | Not set | Powerfront | Sets the warning sent when one more matched message will reach the blocking threshold. |
| Autoblock after (x) matched messages | Settings › General Settings › Censorship | website | 3 messages | Powerfront | Sets how many matched visitor messages trigger automatic blocking. |
| Block for (x) days | Settings › General Settings › Censorship | website | 0, indefinitely | Powerfront | Sets the configured censorship block duration; zero means no automatic expiry. |
| Whitelist Phrases | Settings › General Settings › Censorship | website | Not set | Powerfront | Exempts listed phrases for the selected language. |
| Blocked Phrases | Settings › General Settings › Censorship | website | Not set | Powerfront | Adds language-specific phrases to the system default blocked phrases. |
Limits and good to know
- The threshold counts matched messages, rather than the number of abusive words in one message.
- Censorship applies to visitor content; it is separate from spam classification.
- Masking the visible message does not erase the original moderation evidence. INSIDE retains the original text and matched phrases for censorship records.
- Language-specific lists and warnings need their own review. A phrase permitted in one language is not automatically permitted in every language.
Depends on and works with
View all connections as a list
This feature uses
- Roles and permissionsConfigured by
Censorship setup is restricted to administrative access.
- Supported languagesConfigured by
Supported languages provide the language tabs for phrases and warnings.
- Email templates and ticket notificationsConfigured by
The Censorship email template controls the layout of warning emails.
- Contact opt-in and blockingEnhances
For a linked contact, automatic censorship blocking also blocks that contact's other visitors.
- AI assistance, translation and sentiment reportsRelated
Censorship reports let supervisors review matched messages and blocking evidence.
Used by
- AI guardrail routing and fallback actionsEnhances
Inbound guardrail decisions can use censorship handling.
- AI assistance, translation and sentiment reportsEnhances
Censorship supplies the visitor messages available for review.
FAQ and troubleshooting
Why is an acceptable phrase masked?
Ask Powerfront to check the system list and Blocked Phrases for the relevant language, then add and save an approved exception in Whitelist Phrases.
Why did the visitor receive no warning before being blocked?
Ask Powerfront to check both warning messages, their language versions and the threshold; a threshold of one blocks on the first matched message.
Why are other channels for the same contact blocked?
Automatic censorship blocking applies to the linked contact where one exists. Review the contact's blocked status before unblocking it.