A regular-expression filter configuration for Simply Filter SMS on iOS.
The ruleset automatically identifies almost everything that isn't human while making sure personal texts remain in your Primary Inbox. Download the latest simply.sfsfilters to get started.
- An iPhone running iOS 16.6 or later.
- Simply Filter SMS installed from the App Store.
Apple requires third-party SMS filters to be explicitly granted permission in system settings.
- Open Settings on your iPhone.
- Go to:
- iOS 18 and later: Settings → Apps → Messages → Unknown & Spam
- iOS 17 and earlier: Settings → Messages → Unknown & Spam
- Make sure Screen Unknown Senders (or Filter Unknown Senders) is turned ON.
- Under Text Message Filter (or SMS Filtering), select Simply Filter SMS and tap Enable when asked.
Download the simply.sfsfilters file.
- Safari: Tap Download when asked.
- Files: Save the file in the Files app.
After downloading, import the file into Simply Filter SMS.
- Safari: Tap the Share button and select Simply Filter SMS.
- Files: Open Simply Filter SMS → Filter Tools → Import Filters → select
simply.sfsfilters.
Turn Automatic Filtering (AI) OFF and Smart Filters (All) OFF. The rules handle the filtering, including @ senders, unsafe websites and homograph letters.
Apple's IdentityLookup framework does not permit third-party extensions to delete incoming messages. Instead, incoming SMS, MMS and RCS messages from numbers not saved in your contacts are evaluated upon receipt and routed into dedicated system folders.
- Scams go to Junk.
- Sales and deals go to Promotions.
- Receipts and deliveries go to Transactions.
- Personal messages, codes and bank alerts stay in your Primary Inbox.
Messages from people saved in your Contacts bypass third-party filtering and go to the Primary Inbox.
- Junk: Once a conversation goes to Junk, future messages from that sender will stay there.
- Deleting the conversation: If you delete the conversation, a new message from the sender will be checked again as a new conversation.
- Promotions, Transactions and Primary Inbox: These folders are based on the most recent message.
Important: Do not reply to scam messages. If you reply, future messages from that number will bypass the filter.
- SMS, MMS and RCS: These messages can be checked by third-party filters.
- iMessage: Blue-bubble messages cannot be checked by third-party filters. Use Report Junk for unwanted iMessages.
The ruleset runs on your device. Messages are checked on your device and are not sent to an external server.
If the app says a message should be Junk but it went to your Primary Inbox, check:
- Blue bubble: The message is an iMessage and cannot be checked by the filter.
- Saved contact: Messages from contacts bypass filtering.
- You replied: Future messages from that sender bypass filtering.
- Filter settings: Make sure Screen Unknown Senders and Simply Filter SMS are turned ON under Settings → Apps → Messages → Unknown & Spam.
If none of these apply, turn the filter OFF, wait a few seconds, then turn Simply Filter SMS back ON. You can also close Messages or restart your iPhone.
Bank alerts with “Reply YES or NO” should go to your Primary Inbox. Alerts that only tell you to call an unknown number might be flagged as scams. Always check bank alerts through your bank's official app or the phone number on the back of your card.
Messages like “Hi, is this David?” might not have enough information to identify them as scams. The filter waits for things like links, payment requests, or other scam patterns.
Messages from USPS, UPS, FedEx and DHL using their official websites are allowed through. Delivery messages using unknown tracking websites might still be flagged if they look like scams.
If a real message went to Junk or Promotions:
- Copy the message.
- Remove personal information like names, phone numbers, addresses and account details.
- Open a GitHub issue with the message, the folder it went to, and who sent it.
If a scam message reached your Primary Inbox:
- Copy the message.
- Make sure it has something like a link, phone number, fake brand, or payment request.
- Open a GitHub issue with the message and the scam link or phone number.
Note: Do not report simple messages like “Hey, how are you?” unless they show clear signs of a scam. Blocking every simple greeting would also block messages from real people.
This ruleset is made for everyday use. I'll only update it when major new scam patterns appear. Feedback and contributions are welcome! Know that these rules will not catch every scam out there. Be careful with unknown messages; don't click links or share personal information unless you know the message is absolutely safe. This project uses the MIT License.