This app uses machine learning to intelligently filter your text messages, categorizing them beyond just spam. It offers detailed classifications for financial, order, and weather-related messages, all processed offline on your device for enhanced privacy. Users can contribute to its learning by reporting misclassified messages.
Summarized from real user reviews of 熊猫吃短信2. Based on 234 reviews and a 4.6/5 average, it highlights the 11 positives and 15 negatives reviewers mention most about this Utilities app.
26.7.3
July 31, 2026
更新 AI 知识库到 7 月 20 日。请注意升级 iOS 大版本后可能会重置系统设置里的过滤选项，如果过滤失效，请检查下系统设置哦。
获得最新的垃圾短信拦截机器学习模型
大哥，开发者不知道你订阅时用的 apple id 是什么，既拿不到你的邮箱也拿不到你的手机号，怎么可能会一直登录你的账号。你以为拦截短信会知道你的手机号，实际上 Apple 根本不会分享收件手机号是什么。从头到尾都不知道你是谁，怎么能窃取你账号。
5 issues
Custom rules not syncing across devices
Subscription invalid after changing region
8 requests
Custom rule migration or iCloud sync
Batch import keywords
This app uses advanced machine learning models to filter and categorize your text messages. It goes beyond basic spam detection to identify and sort messages into categories like financial, orders, and weather, all processed locally on your device for maximum privacy.
The app processes all message filtering and categorization offline on your device using CoreML. It explicitly states that it does not upload or collect your text message content. Only anonymized text and your classification are collected if you voluntarily submit message samples to help the app learn.
The app's machine learning models are trained on a vast dataset of anonymized SMS messages reported by users. It learns to identify semantic features of messages to accurately distinguish between normal, transactional, and spam messages without requiring manual keyword configuration.
Yes, you can help improve the app's accuracy by 'feeding' it new message samples within the app. When the app misclassifies a message, you can report it, which helps the machine learning models learn and become smarter over time.
Beyond general spam, the app can identify more granular categories such as financial messages (bank, insurance, utilities), order-related messages (pickup codes, logistics, travel), operator messages, and weather alerts.
