Fournit aux agents IA un accès en lecture seule aux données publiques du Google Play Store, incluant les détails des applications, les résultats de recherche, les avis et les informations sur les développeurs.
Aussi connu sous kryuchenko/google-play-scraper, gpscrape
- Étoiles
- 0
- Outils
- 21
- Téléchargements, 30 jours
- —
- Dernière version
- 0.1.2
- Dernier commit
- 14 nov. 2026
- Licence
- MIT
Qu'est-ce que Google Play Store MCP ?
Fournit aux agents IA un accès en lecture seule aux données publiques du Google Play Store, incluant les détails des applications, les résultats de recherche, les avis et les informations sur les développeurs. Il expose 21 outils, comme play_search_apps, play_get_app, play_get_apps.
Résumé généré à partir du README du projet
À propos du MCP server Google Play Store
Les développeurs peuvent offrir aux agents IA un accès en lecture seule aux données publiques du Google Play Store grâce à ce MCP server. Il permet aux agents de récupérer les détails d'une application, des résultats de recherche, des avis d'utilisateurs et des informations sur les développeurs. Le serveur est conçu pour fonctionner avec des clients locaux tels que Claude Code, Codex CLI et OpenCode via stdio, ou des clients distants comme ChatGPT et l'API OpenAI via streamable HTTP.
Lors d'une utilisation à distance, le serveur nécessite un jeton pour la sécurité, qui peut être transmis via un en-tête d'autorisation ou comme partie d'un chemin URL secret. Il inclut des outils spécifiques tels que play_search_apps pour la recherche par mots-clés, play_get_chart pour accéder aux classements des applications gratuites ou payantes, et play_get_reviews pour récupérer les retours utilisateurs. Pour les opérations de données en masse, comme le catalogage de sitemaps ou l'extraction complète d'avis, le serveur oriente les agents vers l'outil CLI gpscrape distinct. Cette distinction garantit que le MCP server reste léger et réactif pour les tâches interactives des agents.
Extrait du README du projet. Chaque cas d'utilisation est vérifié par rapport à une citation du document.
Comment installer Google Play Store MCP
Formats vérifiés le 10 nov. 2026Ajouter à CursorAjouter à GooseInstaller Google Play Store MCP dans Claude Code
$ claude mcp add gpscrape -- docker run -i --rm ghcr.io/kryuchenko/gpscrape-mcp:0.1.2Installer Google Play Store MCP dans Cursor
~/.cursor/mcp.json
{
"mcpServers": {
"gpscrape": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"ghcr.io/kryuchenko/gpscrape-mcp:0.1.2"
]
}
}
}Installer Google Play Store MCP dans VS Code
.vscode/mcp.json
{
"servers": {
"gpscrape": {
"type": "stdio",
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"ghcr.io/kryuchenko/gpscrape-mcp:0.1.2"
]
}
}
}Installer Google Play Store MCP dans VS Code CLI
$ code --add-mcp '{"name":"gpscrape","type":"stdio","command":"docker","args":["run","-i","--rm","ghcr.io/kryuchenko/gpscrape-mcp:0.1.2"]}'Installer Google Play Store MCP dans Claude Desktop
claude_desktop_config.json
{
"mcpServers": {
"gpscrape": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"ghcr.io/kryuchenko/gpscrape-mcp:0.1.2"
]
}
}
}Installer Google Play Store MCP dans Codex
$ codex mcp add gpscrape -- docker run -i --rm ghcr.io/kryuchenko/gpscrape-mcp:0.1.2Installer Google Play Store MCP dans Codex config.toml
~/.codex/config.toml
[mcp_servers.gpscrape] command = "docker" args = ["run", "-i", "--rm", "ghcr.io/kryuchenko/gpscrape-mcp:0.1.2"]
Installer Google Play Store MCP dans Gemini CLI
$ gemini mcp add -s user gpscrape docker -- run -i --rm ghcr.io/kryuchenko/gpscrape-mcp:0.1.2Installer Google Play Store MCP dans Gemini settings.json
~/.gemini/settings.json
{
"mcpServers": {
"gpscrape": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"ghcr.io/kryuchenko/gpscrape-mcp:0.1.2"
]
}
}
}Outils MCP Google Play Store 21
Détectés depuis le READMEplay_search_appsplay_get_appplay_get_appsplay_get_chartplay_get_reviewsplay_get_similar_appsplay_get_developer_appsplay_get_permissionsplay_get_data_safetyplay_suggest_queriesplay_list_categoriesplay_check_availability
| play_search_apps | — | |
| play_get_app | — | |
| play_get_apps | — | |
| play_get_chart | — | |
| play_get_reviews | — | |
| play_get_similar_apps | — | |
| play_get_developer_apps | — | |
| play_get_permissions | — | |
| play_get_data_safety | — | |
| play_suggest_queries | — | |
| play_list_categories | — | |
| play_check_availability | — |
Afficher les 21 outilsAfficher moins d'outils
invalid_argumentnot_foundnot_listedrate_limitedtimeoutbusycancelledstore_errorinternal
| invalid_argument | — | |
| not_found | — | |
| not_listed | — | |
| rate_limited | — | |
| timeout | — | |
| busy | — | |
| cancelled | — | |
| store_error | — | |
| internal | — |
Configuration
GPSCRAPE_MCP_TOKEN
| Nom | Description | Type |
|---|---|---|
| GPSCRAPE_MCP_TOKEN | — |
Vérifications automatisées
Exécuté le 13 nov. 2026Rapport complet →A passé nos vérifications automatisées. Il ne s'agit pas d'un audit de sécurité : lisez le code et donnez au serveur l'accès le plus restreint dont il a besoin. Fonctionnement des vérifications
- Dépôt source lié et accessibleRéussi
- Licence open source (OSI)Réussi
- Commit ou version publiée au cours de l'année écouléeRéussi
- Nom du package non proche d'un package populaireNon vérifié
- L'endpoint distant utilise HTTPSNon vérifié
- Les outils destructifs sont étiquetésRéussi
- Aucune instruction cachée destinée au modèleRéussi
- Aucun avis critique dans le package ou ses dépendancesNon vérifié
README
Voir sur GitHubgpscrape-mcp
An MCP server that gives AI agents read-only access to public Google Play Store data: app details, search, charts, reviews, similar apps, developer listings, permissions, data safety, search suggestions and per-country availability.
It works with any MCP client. Claude Code, Codex CLI and OpenCode start it as a local process over stdio. ChatGPT, the OpenAI API and other remote clients reach it over Streamable HTTP.
This is a separate Go module, so the root library keeps its zero-dependency
guarantee. The long sweeps (the sitemap catalog, every review of an app) are
left out on purpose and stay in the gpscrape command. The
server tells the agent so, in its instructions and again in the play_get_reviews
tool description (some clients, such as claude.ai and the OpenAI Agents SDK,
do not pass instructions to the model). An agent that can run shell commands
is pointed at gpscrape for bulk jobs if it is already installed. If it is
not, the agent pages with the tools or tells the user the job needs the CLI,
and it is told to ask the user before installing it. The instructions open
with the untrusted-data warning and stay short; every tool description stays
under 1000 bytes, where Codex cuts them for servers installed as plugins.
Install
go install github.com/kryuchenko/google-play-scraper/mcp/cmd/gpscrape-mcp@latest
Requires Go 1.27. Without Go:
- Prebuilt binaries for macOS, Linux and Windows are attached to each
mcp-v*release (the Go module tag ismcp/v*; the release uses a tag without the slash, which the MCP Registry requires in download URLs). - MCPB bundle:
gpscrape-mcp.mcpbon the same release installs in one step in clients that take MCP bundles, such as Claude Desktop. It carries macOS (universal), Linux (amd64 and arm64) and Windows (amd64) binaries. - Container:
docker run -i --rm ghcr.io/kryuchenko/gpscrape-mcpspeaks stdio; for HTTP,docker run --rm -p 8080:8080 -e GPSCRAPE_MCP_TOKEN=<token> ghcr.io/kryuchenko/gpscrape-mcp -http 0.0.0.0:8080. - MCP Registry: listed as
io.github.kryuchenko/gpscrape, so clients that browse the registry can find it.
Every release file and the image carry signed build provenance. Check a download before running it:
gh attestation verify gpscrape-mcp.mcpb --repo kryuchenko/google-play-scraper
gh attestation verify oci://ghcr.io/kryuchenko/gpscrape-mcp:<version> --repo kryuchenko/google-play-scraper
For agents that also have a shell, the
gpscrape skill explains when to reach for the
CLI instead of these tools.
Connect a client
Claude Code
claude mcp add gpscrape -- gpscrape-mcp
Codex CLI: ~/.codex/config.toml
[mcp_servers.gpscrape]
command = "gpscrape-mcp"
OpenCode: ~/.config/opencode/opencode.json or opencode.json in a project
{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"gpscrape": { "type": "local", "command": ["gpscrape-mcp"], "enabled": true }
}
}
Over HTTP: for clients that cannot start a local process
gpscrape-mcp -http 127.0.0.1:8080 # this machine only, no token
GPSCRAPE_MCP_TOKEN=$(openssl rand -hex 32) \
gpscrape-mcp -http :8080 # any interface, token required
The endpoint is http://<host>:8080/mcp. Clients send
Authorization: Bearer <token>. GET /healthz answers ok for probes. The
server will not bind a non-loopback address without a token. Put TLS in front
of it (nginx, Caddy) before exposing it to the internet. Without a token the
server refuses requests whose Host is not local (DNS-rebinding protection),
so a reverse proxy that forwards the public Host needs a token set.
Every method rejects an invalid Origin with 403, even if Fetch Metadata
claims the request is same-origin. Native clients can omit Origin; browser
origins must be HTTP(S) and match the request's public Host.
HTTP is sessionless: no Mcp-Session-Id is issued, and GET/DELETE return 405.
Protocol 2026-07-28 requests stop upstream work when their HTTP request is
cancelled. Legacy HTTP clients' cancellation notifications cannot reach work
running in a different request's temporary session; the per-call timeout
still bounds that work. Stdio cancellation uses the persistent connection.
Give the token in $GPSCRAPE_MCP_TOKEN or in a file with -token-file
(a Docker or systemd secret, for instance). There is no -token flag: a
secret on the command line is visible to every user of the machine in ps.
ChatGPT cannot start a local process and cannot send a token: its custom
apps take OAuth or nothing. Host the server yourself (a VPS, or your own
machine behind a tunnel such as Cloudflare Tunnel) with HTTPS in front, start
it with -secret-path, and give ChatGPT the token as the last part of the
URL:
https://play-mcp.example.com/mcp/<GPSCRAPE_MCP_TOKEN>
Add it in ChatGPT as a custom app (developer mode) with
No authentication: the secret path is the password. It is off unless
-secret-path is given, because a token in a URL ends up in proxy logs and
browser history; any path under /mcp/ answers 404 without it, and so does a
wrong token with it. Use a long random token (openssl rand -hex 32); the
server warns when it is shorter than 24 characters. Keep your proxy from
logging request paths, and rotate the token if the URL leaks. Requests to
Google Play then come from the host's IP, and Google rate-limits data-centre
addresses sooner than a home one.
claude.ai (and Claude Desktop, which shares claude.ai's connectors) does
not need the secret path: its custom connectors can send a fixed header.
claude.ai/customize/connectors → Add custom connector, enter
https://play-mcp.example.com/mcp, choose "No sign in", and under Request
headers add Authorization with the value Bearer <GPSCRAPE_MCP_TOKEN>
(the field takes the value as is, so the Bearer prefix is yours to type).
Claude connects from Anthropic's cloud, not from your device, so the server
must be reachable from the internet as for ChatGPT. claude.ai does not pass a
server's instructions to the model
(anthropics/claude-ai-mcp#93);
the play_get_reviews description repeats what matters, and the tools work
without them. (Anthropic: custom connectors)
For Claude Desktop on one machine there is a simpler way: install
gpscrape-mcp.mcpb from the release page, and Desktop runs the server
locally over stdio.
Remote clients that can set a header point at the URL:
# Codex
[mcp_servers.gpscrape]
url = "https://play-mcp.example.com/mcp"
bearer_token_env_var = "GPSCRAPE_MCP_TOKEN"
// OpenCode
"gpscrape": {
"type": "remote",
"url": "https://play-mcp.example.com/mcp",
"headers": { "Authorization": "Bearer {env:GPSCRAPE_MCP_TOKEN}" }
}
Tools
| Tool | What it returns | Default / max |
|---|---|---|
play_search_apps |
apps matching keywords, with package ids | 20 / 50 |
play_get_app |
one app's store page, including whether it can be installed in that country; response_format: detailed adds description, histogram, developer contacts |
— |
play_get_apps |
concise details for several package ids in one request | up to 20 ids |
play_get_chart |
a chart: top_free, top_paid, grossing, new_free, new_paid, movers_shakers |
50 / 200 |
play_get_reviews |
one page of reviews; next_token for the next page; filter by star rating |
20 / 100 |
play_get_similar_apps |
apps the store lists as similar | 20 / 100 |
play_get_developer_apps |
apps by one developer | 20 / 100 |
play_get_permissions |
Android permissions, grouped | — |
play_get_data_safety |
declared data collection, sharing and security practices | — |
play_suggest_queries |
autocomplete suggestions for a query | — |
play_list_categories |
category ids accepted by play_get_chart |
— |
play_check_availability |
per country: available, region_locked (listed, not installable) or not_listed |
up to 50 countries |
Tools take lang and country (default en, us), with three exceptions:
play_get_reviews takes only lang, play_check_availability takes a list
of countries, and play_list_categories takes nothing. An unknown country code or a malformed language
code is an error, because the store would silently answer with its defaults.
In play_get_reviews, lang selects the language group of the reviews (Google's
grouping, so a few may be in other languages); reviews are the same in every
country. Prices, charts and installability depend on country.
play_search_apps, play_get_chart, play_get_similar_apps and
play_get_developer_apps fetch full details for their
results (one extra batched request per 32 apps), because the store's result
rows carry no developer id, and outside charts no price.
Design notes
-
Names say what they do.
play_+ verb + object, as Anthropic's tool guidance (namespacing) and OpenAI's (action-oriented names) both ask; the prefix also keeps them apart from another store's tools in one session. -
Small answers by default. Lists come back as one-line summaries.
play_get_reviewscuts each text to 400 characters unless you ask fordetailed, and says how many it returned, which filters applied and whether anything was cut. -
Schemas both ways. Input schemas carry
minimum/maximum,enumand array lengths, so a client can check an argument before calling; an omitted number takes its default, an explicit0is out of range. Every tool also publishes an output schema, as OpenAI's MCP guidance asks for structured results, and the server validates each result against it before sending. That roughly doublestools/list(about 17.5 KB, from 9.8 KB); clients with tool search load definitions on demand and pay it only for the tools they use. Results come back asstructuredContentand as the same JSON incontent, because some clients (the OpenAI Agents SDK by default) show the model onlycontent. -
Ids as they come.
app_idtakes a package id or a store link with?id=. -
One contract with the CLI. The records these tools return are the ones in package
records, andgpscrapewrites the same objects one per line, with the same error codes. A filter or parser written against one works on the other;records/testdata/fields.txtpins the field names. -
Third-party text is marked. Review text, reviewer names, developer replies, app summaries, descriptions and release notes come back in fields named
untrusted_*, and the server's instructions tell the agent to treat them as data, not instructions. -
Errors say what to do next, behind a stable code. Every tool error reads
code: what to do. A name instead of a package id points toplay_search_apps; an unknown category points toplay_list_categoriesbefore any request is made; a 404 says to check the id or the country; a block says to wait; a bug that panics returns an error instead of ending the session. The codes:Code Meaning Retry? invalid_argumentfix the arguments no not_foundno such app or page in this storefront another id or country not_listedthe store returned nothing for this id here once, then check the id rate_limitedthe store is refusing this server for now after a few minutes timeoutthe store did not answer in time once busythis server is at capacity shortly cancelledthe client cancelled the call — store_errorany other failure talking to the store once internala bug in gpscrape-mcp no; report it Arguments that break the input schema are rejected by the MCP SDK before the tool runs; those errors start with
validating "arguments". -
Gentle on the store, bounded on the server. Requests to Google Play are throttled and retried honouring
Retry-After. Tool calls run at most-max-concurrentat once, at most four times that many wait for a slot, and any further call is refused at once asbusy. String arguments have a maximum length in the schema.play_check_availabilitygets twice the-timeout. -
Progress for the slow tool.
play_check_availabilitymakes one store request per country and can take two minutes for fifty. A client that sends aprogressTokengets a notification per country over stdio (the stateless HTTP transport answers with one JSON body and has no stream to send them on). -
Cancelled means cancelled. A call the client cancelled returns no result, even if the work behind it finished.
-
Cacheable catalog.
tools/listandserver/discoversayttlMsof an hour: the tool set never changes while the server runs.
Flags
-http addr serve Streamable HTTP instead of stdio
-token-file path read the bearer token from a file (default $GPSCRAPE_MCP_TOKEN)
-secret-path also accept the token as /mcp/<token>, for ChatGPT
-throttle 300ms minimum interval between requests to Google Play
-max-concurrent 4 tool calls running at once
-timeout 60s time limit for one tool call
-v log every tool call to stderr
-version print the version
Logs go to stderr only. In stdio mode stdout carries the protocol.
Conformance
CI runs the official MCP conformance suite
against the HTTP server. Most of its server scenarios call fixtures of the
reference test server (test_simple_text, test_sampling, test prompts and
resources) or features this server does not offer (completion, resource
subscriptions, sessions); conformance-baseline.yml
lists them with the reason. Everything else must pass. To run it locally:
gpscrape-mcp -http 127.0.0.1:3001 &
npx -y @modelcontextprotocol/conformance@0.1.16 server \
--url http://127.0.0.1:3001/mcp --expected-failures conformance-baseline.yml
Questions sur Google Play Store MCP
Comment ajouter Google Play Store MCP à Claude Code ?
Exécutez claude mcp add gpscrape -- docker run -i --rm ghcr.io/kryuchenko/gpscrape-mcp:0.1.2 dans votre terminal. Ajoutez --scope user pour le rendre disponible dans tous les projets.
Comment ajouter Google Play Store MCP à Cursor ?
Ouvrez ~/.cursor/mcp.json et ajoutez le serveur sous mcpServers, avec la configuration Cursor indiquée dans la section d'installation ci-dessus.
Quels outils Google Play Store MCP fournit-il ?
Il expose 21 outils, dont play_search_apps, play_get_app, play_get_apps, play_get_chart, play_get_reviews.
Quel type de données puis-je consulter ?
Vous pouvez accéder aux détails des applications, aux résultats de recherche, aux classements, aux avis, aux applications similaires, aux fiches développeurs, aux autorisations, à la sécurité des données et à la disponibilité par pays.
Comment installer le serveur ?
Vous pouvez l'installer via Go en utilisant go install github.com/kryuchenko/google-play-scraper/mcp/cmd/gpscrape-mcp@latest ou utiliser des binaires précompilés et des conteneurs Docker.
Puis-je l'utiliser avec ChatGPT ?
Oui, mais comme ChatGPT ne peut pas lancer de processus locaux, vous devez héberger le serveur vous-même et fournir le jeton dans l'URL.
Les données sont-elles sûres pour servir d'instructions ?
Non, les textes tiers comme les avis et les descriptions sont marqués comme données non fiables afin que l'agent sache qu'il doit les traiter comme des données et non comme des instructions.