Documentation Index
Fetch the complete documentation index at: https://agencykey.com/docs/llms.txt
Use this file to discover all available pages before exploring further.
Overview
The Sensitive Words feature helps you filter unwanted words or phrases in chats. This protection prevents policy violations and blocks inappropriate content from being sent. AgencyKey includes a default list of restricted words that you can customize to fit your needs.Enabling protection
Use the Protection status toggle to enable or disable word filtering:- Enabled — messages containing restricted words are blocked before sending
- Disabled — no filtering is applied
Managing your word list
The Restricted words section displays all blocked words, one per line. Available actions:- Edit the text field directly to add or remove words
- Click Apply changes to save your modifications
- Click Restore default list to reset to the original word list
Searching words
Use the Search words… field to check if a specific word is blocked:- Word found (green) — the word is in your blocked list
- Word not found (red) — the word is not blocked
Using wildcards
The* symbol allows flexible matching patterns:
| Pattern | Blocks | Does NOT block |
|---|---|---|
| meet | only “meet” | meeting, meetup |
| meet* | meet, meeting, meetup | unmeet |
| *damn | damn, goddamn | damnation |
word*— blocks all words starting with the specified text*word— blocks all words ending with the specified text
What happens when triggered
When a restricted word is detected:- The message is blocked and not sent
- A notification appears explaining the block
- You must remove the restricted word to send the message