> ## 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.

# PFP Review

> Inspect flagged avatars and take action on near-matches from the Dashboard.

PFP Review lets your team visually inspect avatars that closely resemble protected images — and take action directly from the Dashboard.

<Tip>
  The [Hashbot Dashboard](https://dashboard.hashbot.com) is the recommended way to review PFP matches — inspect flagged avatars side-by-side and take action in a few clicks.
</Tip>

### How It Works

1. Run `/pfp-filters review` in any channel where Hashbot is active
2. Hashbot scans your server for avatars similar to any image in your PFP filter list
3. You receive a summary showing the count of pending matches
4. Click the **"Review on Dashboard"** button to open the full review
5. On the Dashboard, inspect each flagged avatar side-by-side with the matched filter image
6. Take action on each match: **Ban**, **Mark Safe**, or **Skip**

<Info>
  These are *near-matches* — they fall below the automatic ban threshold but are still visually similar to known scam images or protected avatars.
</Info>

### Your Reviews Improve Detection

Every time you confirm or dismiss a match, Hashbot refines its detection:

* **Confirmed bans** help Hashbot recognize visual variations (flipped logos, color shifts, overlays)
* **Marked safe** entries teach Hashbot to avoid similar false positives
* Over time, this reduces manual review and improves accuracy

### Dashboard Features

The Threat Review page on the Dashboard provides:

* **Full Scan vs Cached Scan** — run a fresh scan or review cached results
* **Adjustable threshold** — slide between 0.50 and 1.00 to control sensitivity
* **Side-by-side comparison** — member avatar next to the matched filter image
* **Match metadata** — score, user roles, account age, join date
* **Bulk actions** — process multiple matches efficiently

### When to Run Reviews

* **Weekly** as part of routine moderation
* **After applying new PFP templates** — check for new matches
* **After uploading team or brand images** — verify detection is working
* **After a scan** — run `/pfp-filters cleanse` first, then review results

<Tip>
  Review results persist on the Dashboard until you act on them, so you can work through flagged users at your own pace.
</Tip>
