Release notes

What changed in each version. The same notes are in the app under ⚙ → What's new.

v0.9.234 · October 18, 2026

The "Not a concern?" pop-up now says plainly that a shared false-alarm report includes what Privacy Watch saw the app do (which access, how often, when, how much it sent), not only facts about the app.
Fixed: the app you were using on screen could be counted as "not in use", so its web lookups, uploads and sensor use were wrongly shown as happening in the background.
Fixed: apps with "face" in their name, such as Facebook and Messenger, were explained as "face unlock". Only real face-unlock and biometric components get that explanation now.
Fixed: history older than 30 days is now also removed from the file on your phone while monitoring keeps running, not only when the app restarts.
Fixed: another app's name can no longer add fake entries to the error log that goes to support.
When Android blocks an app from reading something (because you turned the permission off or applied a fix), Privacy Watch now shows it as "blocked by Android (not allowed)" on the timeline. It no longer counts as an access, a possible policy breach or "not in use", and what the app sent afterwards isn't linked to it. Found by the automated support-log review.
Apps that use something you allowed for an optional feature, then talk to their own servers, are no longer flagged red when nothing points to ads and their privacy policy is graded A or B. Example: Signal checking your contacts to find friends who use Signal. They show green with "an optional feature you allowed" (yellow if the policy is graded C). Access you didn't allow, ad-network contact, or a D/F policy still turns it red.
While monitoring, web-address lookups now go only to your own network's DNS servers. Privacy Watch no longer falls back to Cloudflare's public DNS server when yours doesn't answer. Phones on mobile networks that use only newer (IPv6) DNS servers are covered too.

v0.9.162 · October 16, 2026

Your Privacy Watch history no longer moves to a new phone when you copy your old phone over during setup. It was already kept out of cloud backups; now it stays on the phone where it was recorded.
Privacy-policy grades now follow updated rules, so many apps' grades changed: – Getting data about you from other companies is still shown, but no longer lowers the grade. Selling your data or sharing it for ads is what counts most. – Handing data to authorities only counts against an app when its policy allows it on request or "as it believes appropriate", without a warrant, court order or subpoena. Complying with legal process, or reporting an emergency, doesn't count. – An app that sells or shares your data for ads, but lets everyone opt out, is graded C at worst. – A Google Play label that says no data is shared now forces "D at best" only when the policy allows giving data to another company for its own use (ads, sale, data brokers). Service providers and anonymized data don't count, matching Google's own rules.
When one privacy policy covers many apps (Google, Amazon, Microsoft and others), what it allows may not apply to the app you have. Its Play label differences still count, but no longer force a D, and the app card and reports now say so: "This privacy policy covers N apps MCPOTA reviewed, so what it allows may not apply to this app."
Dashboard → 📜 Privacy policies has a search box to find an app by name, and filters: All, A–B, C or D–F. Tapping a grade tile filters too.
More apps have a privacy-policy review. When a company's site blocks automated reading, MCPOTA reviews the Internet Archive's dated copy of the same page, and the app card says so.
🛡 End-of-service apps are flagged red as a security risk (not a privacy one), with a "🛡 Security risk" tag and Uninstall as the fix. When an app you installed is no longer on Google Play (MCPOTA checks the store), the Play Store sends it no updates, security fixes included, so its card says so and recommends uninstalling it (with a supported replacement first if you still need what it does).
🛡 An app that's off Google Play but was updated in the last 12 months is yellow instead: it may be a work or private app that still gets updates privately (for example through your employer's device management).
🛡 An app installed from an app store that hasn't been updated in 2 years or more is flagged yellow as a possible security risk: it may no longer be maintained. Apps that came with your phone aren't flagged this way. Both checks use facts on your phone and MCPOTA's own store check; nothing is looked up from your phone.

v0.9.159 · October 16, 2026

The troubleshooting log (and Send logs to MCPOTA support) now includes the full list of apps on your phone: package names, whether each came with the phone, and which store installed it. No app content is included. The warning you see before sending says so.

v0.9.156 · October 16, 2026

⚖ "Possible policy breach" is now flagged only for promises the phone can actually check: a policy that says the app never accesses something (or has no ads or trackers). Promises that depend on your consent, a feature being on, or what a company does on its servers are still listed, but never flagged.
🔔 Alerts for every sensitive background access. Privacy Watch now alerts you when an app you installed accesses your microphone, camera, location, contacts, texts, call log, calendar, photos or files, clipboard, or scans nearby Wi-Fi/Bluetooth while you're not using it. Before, only the microphone, camera, clipboard and photos alerted.
📤 With advanced access, it also alerts when an app sends 100 KB or more within 5 minutes after such an access: "sent 6.0 MB within 5 minutes of accessing your precise location". It sees how much was sent, never what.
⚖ When an app MCPOTA has reviewed does something its privacy policy says it doesn't, the alert says so: "Possible policy breach: accessed your microphone while closed despite its privacy policy saying it doesn't". This is MCPOTA's comparison of what was observed with the policy's own words, not a finding about the company's conduct.
Access that's expected for that app doesn't alert: your texting app reading texts, your phone app using the call log, your keyboard reading the clipboard, your navigation app using location, a backup app reading photos, a watch or fitness band scanning Bluetooth, and for reviewed apps anything MCPOTA's review marks as expected. Apps that came with the phone don't alert.
One notification per app, updated in place, with the most serious thing first (possible policy breach, then data sent, then access) and the latest five items with their times. It sounds only when something new and more serious appears, and at most once per app every 15 minutes; otherwise it updates quietly. Tapping it opens the app's timeline at the latest event, and Back goes to the app's card.
New ⚙ → Notifications → Alerts: a switch for each kind (Microphone, Camera, Location, Contacts, Texts, Call log, Calendar, Photos & files, Clipboard, Wi-Fi & Bluetooth scans, Data sent after an access, Possible policy breaches). All are on to start. The notification channel is now called "App activity alerts"; your Android settings for it carry over.
Alerts are made on your phone; nothing about them is sent anywhere.

v0.9.153 · October 15, 2026

📤 Access, then data sent. With advanced access, Privacy Watch now checks every 30 seconds while monitoring how much each app has sent, so it can tell when an app sends data shortly after accessing your microphone, camera, location, contacts, texts, call log, calendar, photos, clipboard or nearby Wi-Fi/Bluetooth. It records how much each app sends, never what it sends.
An app's card says so when it happens while the app is closed: "📤 Sent 2.1 MB within 5 minutes of accessing the microphone while closed (Thu 15:44)." This is the strongest sign Privacy Watch shows. For reviewed apps, an access the app isn't expected to make followed by 100 KB or more sent turns it red. For apps without a review, 1 MB or more sent after a background access to your microphone, camera, location, contacts, texts, call log, calendar, photos or clipboard turns it at least yellow. Expected access (a navigation app syncing a trip) gets the line but no colour change.
Each app's timeline has new "⬆ Sent 2.1 MB" rows (marked "Not in use" when the app wasn't open), and each access shows what followed it: "→ sent 2.1 MB and looked up ad.doubleclick.net (tracker) within 5 min". The summary shows how much it sent, measured every 30 seconds. Without advanced access, the timeline says per-access amounts need it, and Android's background totals (about every 2 hours) are shown as before.
A privacy-policy promise not to upload in the background is also compared with these 30-second amounts: "sent 2.1 MB while closed (last Thu 15:44)".
Both reports include it: the 📤 line on app cards with why it matters, and the technical report's access table and per-app timelines show how much was sent after each access. The dashboard's Access → network page includes data sent too.
A finding you marked as not a concern shows again if 📤 appears for the first time, since that's new behaviour.

v0.9.147 · October 15, 2026

⚖ What an app does, compared with what its privacy policy promises. For apps MCPOTA has reviewed, Privacy Watch now checks the specific promises their privacy policy makes (for example "The app does not access your microphone") against what it observed on your phone. If they don't match, the app's card says so in the policy's own words: "⚖ Possible policy breach: accessed your microphone while closed (last Wed 14:12) despite its privacy policy saying: “The app does not access your microphone and never records audio.”" This covers location, Wi-Fi/Bluetooth scans, microphone, camera, contacts, texts, call log, photos and files, clipboard, sending data while closed, and contacting ad networks.
A possible breach moves the app one colour worse (green to yellow, yellow to red), once, however many promises it doesn't match. On Summary, the red card shows the first such line.
Each app's timeline lists "Promises its policy makes" under Privacy policy, with ⚖ and what was observed next to any promise this period's behaviour doesn't match. The technical report's policy section lists them too, with the same ⚖ marks, and both reports explain why a possible breach matters.
Only promises whose quote was found word-for-word in the policy are used. The comparison is MCPOTA's analysis of what was observed against the policy's own words, not a finding about the company's conduct, and it happens entirely on your phone using the policy database built into the app.

v0.9.136 · October 15, 2026

Safer fixes for apps that came with your phone: Privacy Watch no longer offers to disable a system app unless it's on a list of system apps known to be safe to turn off. Turning off the wrong one can break calls, Wi-Fi or the home screen. For other system apps you can still remove their permissions.
Texts, call log and contacts are now shown separately, and calendar access is new. Each has its own line on an app's card, its own filter on Details, and its own Fix it ("Block calendar access" is new). Apps that aren't calendar, mail or work apps are flagged when they use your calendar while closed. Your existing history is sorted into the new groups automatically.
Location now says whether an app accessed your precise location or only your approximate location, on its card, its timeline and in reports, alongside what you allowed.
An app's "What you agreed to" card explains that apps can read motion sensors (accelerometer, gyroscope) without asking, and that Android doesn't record which app did, so Privacy Watch can't show it.
One retention period: Privacy Watch keeps its records for up to 30 days. The error log now keeps 30 days instead of 14, and each log can hold more before trimming. Old logs, reports and the update history are cleared even if nothing new is written. Fixes are kept until you undo them, so Undo always works. Support logs you send us are deleted within 30 days instead of 90.
Clearer wording: findings say what was observed, such as "accessed the microphone" or "sent 12 MB while closed" (Privacy Watch sees how much, never what), and "contacted an ad network within 5 minutes of accessing your location". Policy grades are labelled as MCPOTA's analysis of the policy's text, and "How it could be misused" says it describes what this kind of access makes possible, not something the app was seen doing.

v0.9.133 · October 15, 2026

Informed consent on every finding. Each app's card now says what it did, what you agreed to and what the app says it needs, and whether they match: "🔐 You allowed the microphone only while using the app. It used it while closed through a background service." If an app used the microphone, camera or location while closed when you only allowed it while using the app, the card starts with "⚖ Used outside what you agreed to" and turns red (microphone or camera) or yellow (location).
Apps that hold a role you chose are explained: your texting app reading texts, your phone app using the call log, your assistant using the microphone, your keyboard reading the clipboard. These aren't counted against them.
Each app's timeline has a new "What you agreed to" card: every permission it has and how far it reaches (Allowed all the time, Only while using, Ask every time, Not allowed, Limited photos), special access like reading your notifications, roles, the background work it declares, its store category and where it was installed from. "Change permissions" opens its Android settings. All of it is read on your phone; nothing goes online.
Both reports include the same consent lines on each app.

v0.9.121 · October 14, 2026

Every notification now opens the screen it's about. A microphone, camera, photo or clipboard alert opens that app's timeline at the exact moment it happened, highlighted, and Back takes you to the app's card on the main screen. Pairing notifications open Advanced setup.
📜 Privacy policies count when you decide whether to keep an app. The dashboard has a new Privacy policies widget: how many apps on your phone have policies graded A or B, C, and D or F (in our analysis, D and F leave you exposed). Tap it for every reviewed app, worst first, with its grade and a one-line summary; tap an app for its timeline.
The main screen says when apps on your phone have privacy policies graded D or F, and takes you to that list.
Both reports have a Privacy policies section: what a policy decides, the grade counts, and each reviewed app with its grade and what its policy means in one sentence.
If an update changes an app's policy grade, Privacy Watch tells you for 30 days: "📜 Policy grade changed C → D on Oct 14", on the app's card, its policy, the policy list and the reports.
History now goes back 30 days instead of 7. The dashboard and reports have a new "Last 30 days" option, with charts shown per day. Uninstalled apps stay in the history for 30 days too.
Email report now always opens your email app with the report attached, never a chat app.
A policy re-graded under MCPOTA's updated scoring no longer shows as "Policy grade changed"; only real policy changes do.
7- and 30-day views no longer turn apps yellow just for adding up more days of data, and an app's timeline opened from the dashboard covers the same period.
Chart bars now start on the hour, the 6-hour mark or midnight, so labels match their bars. The dashboard and reports load faster over 30 days.

v0.9.120 · October 14, 2026

Apps whose privacy policy is graded D or F are never green: if they did anything while closed they're at least yellow, and any unexpected access makes them red. The card says "📜 Our analysis of its privacy policy: grade D (leaves you exposed)".

v0.9.115 · October 14, 2026

New red / yellow / green grading: would an average person expect this app to do this? For reviewed apps, Privacy Watch now knows what each app is for and what it needs while closed. Green: it only uses what you'd expect. Yellow: it uses what you'd expect but contacts ad networks or its privacy policy allows advertising use, or it does something unexpected but minor. Red: it uses something it has no reason to (mic, camera, location, contacts, texts, photos, clipboard), or does something unexpected and is linked to advertising in either of those ways. Each card says why, in plain words.
Analytics and crash reporting on their own no longer make an app yellow; contacting advertising networks does.
Fixed: email buttons could open Google Chat instead of Gmail ("Chat not enabled for this account") and nothing was sent. They now always open an email screen.
⚙ → Send logs to MCPOTA support: after explaining what's in it, opens your email with the full troubleshooting log attached. Nothing is sent until you tap Send.

v0.9.109 · October 14, 2026

📜 Privacy policy intelligence: for apps MCPOTA has reviewed, cards, timelines and both reports now show the policy's grade (A–F) and the exact clauses that let the company do something with what the app was seen doing — for example, a location finding shows the clause that allows sharing location with advertisers. The technical report lists the policies of every reviewed app on your phone. The reviews ship inside the app, so nothing is looked up online.
Camera use is now caught on phones that don't record it in Android's access log (seen on Samsung with Android 16).
Fix it: new "Block all internet" (Android 14+) cuts an app off completely, even while it runs in the foreground; red apps that can see all your photos now get "Block photo & video access" too, because Android doesn't record background photo reads. Fixes already applied are no longer offered again.
The fix check now warns only when Android clearly reports a fix was undone, and a fix that Android doesn't support shows an error instead of looking applied.
Reset for testing now really clears history: activity from before the reset no longer comes back after a restart, and timelines show data use correctly afterwards.
Summary keeps Undo for fixed "worth a look" apps.

v0.9.91 · October 13, 2026

⚙ → Reset for testing: undoes every fix and clears findings, history and hidden items, so you can check detection from a clean start.

v0.9.90 · October 13, 2026

Brighter app icon: the cyan MCPOTA emblem on a lighter blue glow, so it stands out on dark and black home screens.

v0.9.89 · October 13, 2026

Fixed: mic, camera and other access used while you were in the app (for example dictating into it) could be counted as "while you weren't using it". Privacy Watch now uses Android's own record of whether the app was on screen, and the access must also end while you're still in the app. Anything that keeps going after you leave is flagged ("continued after you left the app"), and accesses still running are re-checked until they stop.
Timelines now say whether the app was on screen, in the background, or using a foreground service at the time.

v0.9.84 · October 13, 2026

New MCPOTA look: new app icon, the MCPOTA emblem in the app header and on every report, and the MCPOTA motto, Arx Dominii Privati ("stronghold of private dominion"), in About, Settings and report footers.
A short MCPOTA emblem animation plays when the app starts (silent; tap to skip). It plays in full once after each update, then briefly.

v0.9.83 · October 13, 2026

The one-tap report buttons are now at the top: under the verdict on Summary, and above Status on Details.

v0.9.80 · October 13, 2026

New name: MCPOTA Privacy Watch. The home-screen icon still reads Privacy Watch, and the app header shows MCPOTA above the name.

v0.9.79 · October 13, 2026

Tabs renamed: Simple is now Summary, and Advanced is now Details, to match the executive summary and the detailed technical report (and to avoid confusion with "advanced access").

v0.9.78 · October 13, 2026

One-tap reports: "Email me this summary" at the bottom of Simple, and "Technical report" at the bottom of Advanced. The full 📄 Reports screen stays on the Dashboard.

v0.9.77 · October 13, 2026

⚙ now opens a full Settings screen, grouped into Plan, Appearance, Protection, Help and Data.
Theme is one Auto / Light / Dark switch.
Reports moved to a 📄 Reports button on the Dashboard.

v0.9.75 · October 13, 2026

About screen in ⚙: version, plan, links and open-source licences.
What's new (this page) in ⚙.
Logs now record each update, to help with troubleshooting.
Fix check now covers fixes made before v0.9.72.

v0.9.72 · October 13, 2026

Fix check: Privacy Watch notices when an app update, reinstall or system update undoes a fix, and offers Re-apply or Leave it.
Trial: 7 days of full monitoring with one free one-tap fix. Fix it, reports and 7-day history are part of the full version.

v0.9.65 · October 13, 2026

Findings now show only what apps actually did: data sent, background access and trackers.
New 🛡 Settings audit, kept separate: screen readers, notification readers, device admins, keyboards, overlays, contactless-card apps, certificates, hidden apps and screen recording.
"How it could be misused": for every finding, what a company and what an attacker could do with it.
Report a false alarm: hide a finding and optionally send the details to support. Hidden findings can be shown again from ⚙.
Music and video apps scanning for speakers are amber, not red. Social apps turn red for background data only above 5 MB sent.

v0.9.50 · October 12, 2026

New top bar on every screen: Simple, Advanced, Dashboard and ⚙.
Settings menu with Dark/Light mode, Reports, Banking mode and Logs.
PDF reports: faster, smaller, with a progress window while they build.
Executive and technical reports by email, or downloaded.