11 new Play reviews, 4 at one or two stars
Who replies to the 4 low-star reviews?
11 new reviews since yesterday morning, averaging 3.4 stars. 4 are low ratings, and 3 of those mention crashes on version 5.8.0.
Every morning: new Google Play reviews with rating, device and app version, low ratings first, and a question on who replies.
Reads
You get
The card on the right is a sample. Yours shows your own numbers and updates in place every run.
What lands on your board
Who replies to the 4 low-star reviews?
11 new reviews since yesterday morning, averaging 3.4 stars. 4 are low ratings, and 3 of those mention crashes on version 5.8.0.
A one-star review answered the same day often turns into a three-star update; one answered a month later rarely does. The Play Console shows reviews well but doesn't remind anyone, and the reviews API only reaches back a week, so a missed week is gone from the automation for good.
Every morning the agent collects the reviews written or changed since the last run, puts one and two stars first with their device and app version, and drafts short replies. Support and product leads see whether a bad version is behind the ratings, and answer from the card.
Call GET https://androidpublisher.googleapis.com/androidpublisher/v3/applications/{packageName}/reviews?maxResults=100, paging with token. It returns production reviews with text that were created or changed in the last 7 days.
Keep reviews whose comments[0].userComment.lastModified is after the last run (saved in a small JSON file) and that have no developerComment yet. Read starRating, text, appVersionName, androidOsVersion, deviceMetadata and reviewerLanguage, allowing for missing fields.
Put 1 and 2 stars first. Group complaints that repeat on the same appVersionName, and draft a plain-text reply of at most 350 characters for each low rating, in the reviewer's language.
Post the list as an ask. For each reply you approve, call POST .../reviews/{reviewId}:reply with replyText. Each reply posts publicly under the review, so send only the ones that were approved.
With BRIEFKIN_TOKEN set, the agent sends the result as one card that updates in place every run. Without it, you get the report in the chat.
Pick your agent, where the data comes from and how to schedule it; the commands below follow.
One command. It puts the SKILL.md in ~/.claude/skills and in ~/.agents/skills, which Codex, Cursor, Gemini CLI, Copilot CLI and OpenCode read too.
shellcurl -fsSL https://briefkin.dev/skills/google-play-reviews-to-answer/install | shAsk Claude Code:
promptRun the google-play-reviews-to-answer skillOr paste the prompt instead of installing:
promptUsing the Google Play Developer API (reviews.list for package <package>), list the reviews created or changed since yesterday that have no developer reply. Give the star rating, text, device, Android version and app version for each, one and two stars first, and point out complaints that repeat on the same app version. Draft a short, plain-text reply (at most 350 characters) for each low rating in the reviewer's language. Then ask me who replies. Post no reply until I approve it.Daily at 9:00, from cron on any machine where Claude Code is signed in:
crontab0 9 * * * claude -p "Run the google-play-reviews-to-answer skill"Add it with crontab -e. A laptop asleep at that time runs it when it wakes if you use launchd or Task Scheduler instead; GitHub Actions works for the API route.
Connect Briefkin once with curl -fsSL https://briefkin.dev/install | sh and every run lands on your board as a question for you. How setup works
SKILL.md---
name: google-play-reviews-to-answer
description: "Every morning: new Google Play reviews with rating, device and app version, low ratings first, and a question on who replies. Use when the user asks for this, or when it runs on a schedule."
---
# Google Play new reviews to answer
Every morning: new Google Play reviews with rating, device and app version, low ratings first, and a question on who replies.
A one-star review answered the same day often turns into a three-star update; one answered a month later rarely does. The Play Console shows reviews well but doesn't remind anyone, and the reviews API only reaches back a week, so a missed week is gone from the automation for good.
Every morning the agent collects the reviews written or changed since the last run, puts one and two stars first with their device and app version, and drafts short replies. Support and product leads see whether a bad version is behind the ratings, and answer from the card.
## Two ways to get the data
Use the API or MCP route when you have an MCP server, a CLI or an API token for the app. Otherwise, if you can drive a browser the user is signed in to (Claude in Chrome, a Playwright or Chrome DevTools MCP server), use the browser route. Say which one you used.
## Route A: API or MCP server
### What you need
- A Google Cloud service account added in Play Console (Users and permissions) with the Reply to reviews permission, the Google Play Android Developer API enabled, and its key in `GOOGLE_APPLICATION_CREDENTIALS`.
- The app's package name.
If something is missing, say what and stop; do not invent numbers.
### Steps
1. **List recent reviews** (Google Play Console): Call `GET https://androidpublisher.googleapis.com/androidpublisher/v3/applications/{packageName}/reviews?maxResults=100`, paging with `token`. It returns production reviews with text that were created or changed in the last 7 days.
2. **Keep the new ones**: Keep reviews whose `comments[0].userComment.lastModified` is after the last run (saved in a small JSON file) and that have no `developerComment` yet. Read `starRating`, `text`, `appVersionName`, `androidOsVersion`, `deviceMetadata` and `reviewerLanguage`, allowing for missing fields.
3. **Sort and draft**: Put 1 and 2 stars first. Group complaints that repeat on the same `appVersionName`, and draft a plain-text reply of at most 350 characters for each low rating, in the reviewer's language.
4. **Ask, then reply** (Google Play Console): Post the list as an ask. For each reply you approve, call `POST .../reviews/{reviewId}:reply` with `replyText`. Each reply posts publicly under the review, so send only the ones that were approved.
## Route B: the browser
### What you need
- Signed in to Google with a Play Console user who can view app information and has the **Reply to reviews** permission on the app.
If something is missing, say what and stop; do not invent numbers.
### Steps
Start at https://play.google.com/console/. The user is signed in already; never type passwords or codes, and ask if a sign-in page appears.
1. **Open the app's reviews**: Pick the developer account if asked, open the app, and in the left menu go to **Ratings and reviews > Reviews**.
2. **Filter to new, unanswered reviews**: Set the filters above the list: **Reply state** to `Unreplied`, **Date** to the last day (or since the last run), and leave the rating open. Sort by **Most recent**.
3. **Read the low ratings first**: Each review shows its stars, text, device, Android version and app version. Note everything with 1 or 2 stars first, then the rest, and the average of the batch.
4. **Group repeated complaints**: Where several low ratings name the same problem on the same app version, say so, and check **Android vitals** for that version before drafting a reply.
5. **Draft, ask, then reply**: Draft a plain-text reply of at most 350 characters per low rating in the reviewer's language. Report the list and ask who replies. After approval, click **Reply** under the review, paste the text and **Send**.
Note: The console shows all reviews, not just the last 7 days the API returns, and it includes translated text with the original on hover. Star-only ratings without text do not appear in the Reviews list; they count in Ratings.
## The report
- **New reviews** since yesterday, with the average rating.
- **Low ratings first**: stars, text, device, Android version and app version.
- **Repeated complaints** by app version.
- **Draft replies** for the low ratings, and the ask: who replies.
## Post it to Briefkin
When `BRIEFKIN_TOKEN` is set, or `~/.config/briefkin/env` exists (the installer from https://briefkin.dev/docs writes it), also send the result to the user's Briefkin board as one card. Keep the `item_key` the same on every run so the card updates in place, use a new `event_id` per run, and put today's real numbers in place of the example ones. If the request fails, say so once and carry on.
```bash
[ -n "${BRIEFKIN_TOKEN:-}" ] || . ~/.config/briefkin/env
curl -sS -X POST "${BRIEFKIN_URL:-https://briefkin.dev}/v1/events" \
-H "authorization: Bearer $BRIEFKIN_TOKEN" \
-H "content-type: application/json" \
-d @- <<'JSON'
{
"v": 1,
"event_id": "google-play-reviews-to-answer/<date of this run, e.g. 2026-10-16>",
"item_key": "google-play-reviews-to-answer",
"project": "Acme Notes for Android",
"source": "Google Play new reviews to answer skill",
"kind": "action",
"title": "11 new Play reviews, 4 at one or two stars",
"summary": "**11 new reviews** since yesterday morning, averaging **3.4 stars**. **4** are low ratings, and 3 of those mention crashes on version 5.8.0.",
"action": {
"request": "Who replies to the 4 low-star reviews?",
"options": [
"Send the drafts",
"Support replies",
"I will reply",
"Skip today"
]
},
"widget": [
{
"type": "Table",
"props": {
"columns": [
"Stars",
"Review",
"Device",
"Version"
],
"align": [
"right",
"left",
"left",
"left"
],
"rows": [
[
1,
"\"Crashes every time I open a project since the update.\"",
"Pixel 8, Android 15",
"5.8.0"
],
[
1,
"\"Lost my offline notes after updating.\"",
"Galaxy S23, Android 14",
"5.8.0"
],
[
2,
"\"Crashes when I attach a photo.\"",
"Redmi Note 12, Android 13",
"5.8.0"
],
[
2,
"\"Sync is slow on mobile data.\"",
"Galaxy A54, Android 14",
"5.7.2"
],
[
4,
"\"Love the new widgets, wish they were resizable.\"",
"Pixel 7, Android 15",
"5.8.0"
],
[
5,
"\"Best notes app I have used.\"",
"OnePlus 12, Android 15",
"5.8.0"
]
]
}
},
{
"type": "Callout",
"props": {
"text": "3 of the 4 low ratings mention crashes on 5.8.0. Check Android vitals for that version before you reply.",
"tone": "warn"
}
}
]
}
JSON
```
Widgets come from the Briefkin catalog (https://briefkin.dev/v1/catalog). Without a token, just give the user the report.
## Run it on a schedule
Suggested: Daily at 9:00 (`0 9 * * *`). From cron, with the agent signed in on that machine:
```bash
0 9 * * * claude -p "Run the google-play-reviews-to-answer skill"
# Codex: codex exec, Gemini CLI: gemini -p, Copilot CLI: copilot -p, OpenCode: opencode run
```
Or use the agent's own scheduler (Claude Code routines, Codex automations, Claude or ChatGPT scheduled tasks): https://briefkin.dev/skills/google-play-reviews-to-answer#schedule
Source: https://briefkin.dev/skills/google-play-reviews-to-answer
A Google Cloud service account added in Play Console (Users and permissions) with the Reply to reviews permission, the Google Play Android Developer API enabled, and its key in GOOGLE_APPLICATION_CREDENTIALS.
The app's package name.
GOOGLE_APPLICATION_CREDENTIALS at its key and set PLAY_PACKAGE_NAME. Calls use the https://www.googleapis.com/auth/androidpublisher scope, and the reviews API allows 200 reads an hour per app.Grant only the app permissions the report needs: View app information for vitals and reports, and Reply to reviews, which Google's reviews guide asks for. Leave release, pricing and store listing permissions off.
The reviews API returns only reviews with a comment that were created or changed in the last 7 days, and leaves out star-only ratings. That is why the skill runs daily. For older reviews, use the reviews export in Play Console.
Up to 350 characters of plain text; HTML tags are stripped. Replies post publicly under the review, and the reviewer gets a notification, so the skill always waits for your approval.
No. The skill runs with your own agent and gives you the report in the chat. With a Briefkin token set, each run also posts it to your board as a asks you, so you can glance at it instead of asking again.
Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot CLI, OpenCode, Windsurf and Amp all read SKILL.md files. The install command puts it in ~/.claude/skills and ~/.agents/skills, where they look. ChatGPT, the Claude app and Grok take the prompt instead. Pick your agent under "Install and run it" and the commands change to match.
Yes. Switch the data source to "Your browser" and the steps change to the Google Play Console pages themselves: the agent opens play.google.com/console in a browser you are signed in to and reads the numbers off the page. Claude Code, Codex, Cursor, Claude app and ChatGPT can drive a browser out of the box; Gemini CLI, Copilot CLI and OpenCode need a Playwright or Chrome DevTools MCP server first. It is slower than the API and needs you signed in, so it suits a report you run by hand or from a scheduler on your own computer.
Skills are free to use with any agent. Briefkin is where their results land: one board for every report, number and question your agents send, with a ping when something needs you.
Free during the open beta. No card.