> ## Documentation Index
> Fetch the complete documentation index at: https://hashbot.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Phrase

> Phrase filters are straightforward and effective for catching impersonators using specific, case-sensitive text in their usernames or nicknames.

If a user's name contains the exact phrase (case-sensitive), Hashbot will take action.

### How It Works

When you add a phrase to your blocklist, Hashbot checks every new name for that **exact case-sensitive substring**.

#### Example:

`/name-filters add phrase Pascal`

This will match:

* `Pascal`
* `Sir Pascal`
* `Pascalou`
* `Pascal the Great`

But it will **not** match:

* `pascal`
* `Pasca`
* `pascalou`
* `pascal 123`

<Warning>
  **Capitalization matters** — phrase filters are case-sensitive.
</Warning>

### Real-World Example

Impersonators often swap visually similar characters, such as replacing a lowercase **"l"** with an uppercase **"I"**.

#### Example:

If your team member's name is **Signal**, a scammer might use:

* `SignaI` (capital "I" instead of lowercase "l")

To catch this, add:

`/name-filters add phrase SignaI`

Hashbot will flag any username containing this exact string — blocking the visual trick.

### Best Practices

* Use phrase filters for **direct impersonation terms** and known scam words
* Target visual deceptions of team member names
* Add both uppercase and lowercase variants if you want to catch all cases
* Combine with **Fuzzy Mode** (Premium) for protection against Unicode-based spoofing
* Use **regex filters** with `(?i)` for case-insensitive matching when needed

<Tip>
  See how "SignaI" (with a capital `I`) looks like "Signal"? Phrase filters catch this kind of visual trickery. For broader case-insensitive matching, use a regex filter like `(?i)signal` instead.
</Tip>