A real Android phone in the cloud for any MCP agent. Each device is built as a specific handset (Pixel, Galaxy, Xiaomi, Motorola, OnePlus and 150+ models, Android 12-17) with the same build, sensors, cameras, codecs and GPU as the real hardware, Google Play inside, and its own residential, mobile, ISP or datacenter network exit in the country you choose - apps see a phone, not an emulator. Six tools: observe (screenshot with numbered elements + UI tree), act (tap by number, swipe, type in any language, apps, links, intents, settings, clipboard, notifications), inspect (notifications, clipboard, app versions, crashes, health without touching the screen), snapshots (reset every run to the same state, clone one prepared phone into many), recipe (record a flow once, replay it without a model), devices. Local stdio server or hosted endpoint at https://api.clousd.com/mcp with your key. Built for agent evaluation and training, app testing on real models, account operations and any task that needs a phone in a specific country. Docs: https://clousd.com/docs/ - Early access: https://clousd.com/agents/
Score is being calculated.
Do one thing on the phone, then wait until the screen settles. Prefer element numbers from observe over
coordinates; tap_text taps the element whose text contains text. Typing goes into the focused field (tap it
first) and works in any language. Returns what happened; "the screen did not change" means the action probably
missed - observe again.
| Name | Required | Description | Default |
|---|---|---|---|
| x | No | x in screenshot pixels (tap, long_press, swipe start) | |
| y | No | y in screenshot pixels | |
| ms | No | swipe / long_press duration in ms | |
| x2 | No | swipe end x | |
| y2 | No | swipe end y | |
| text | No | type: the text; key: home|back|recents|enter|tab|del|menu; scroll: down|up; open_app/close_app: package; open_url: https://...; tap_text/wait_text: text on screen; settings: wifi|bluetooth|display|sound|apps|location|about|battery|...; intent: a URI to view (geo:, tel:, https:...) or JSON {action,data,package,component,extras}; clipboard_set: the text | |
| action | Yes | one of: tap long_press swipe scroll tap_text wait_text type key open_app close_app open_url installed intent settings clipboard_set notifications_open notifications_close notifications_clear | |
| device | Yes | Device name from devices (op=list), e.g. dev_04f2 | |
| element | No | element number from the last observe (tap, long_press) | |
| timeout | No | wait_text: seconds to wait, up to 120 |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Phones this API key can use.
op=list: name, state (running / stopped / starting), model, Android version and network exit of each phone.
op=start / op=stop with device: start a stopped phone (waits until it has booted, one to three minutes) or stop
one (a stopped phone keeps its state and costs nothing).
| Name | Required | Description | Default |
|---|---|---|---|
| op | No | list (default) | start | stop | list |
| device | No | device name, for start/stop |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Read the phone without touching the screen. notifications: the shade as a list (app, title, text, time). clipboard: the clipboard text. app: installed version, install/update time, installer, whether it runs or is on screen. crashes: recent app crashes, native crashes and ANRs with the first lines of the stack. health: running, booted, network exit and the latest automatic check of the phone. installed: packages of apps that can be opened.
| Name | Required | Description | Default |
|---|---|---|---|
| what | Yes | notifications | clipboard | app | crashes | health | installed | |
| device | Yes | Device name from devices (op=list), e.g. dev_04f2 | |
| package | No | app: the package to describe; crashes: only this package |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Look at the phone: a fresh screenshot, the app on screen and (with_text) every labelled or tappable element with a number. Act on an element by its number (act element=N); coordinates, when needed, are in this image's pixels.
| Name | Required | Description | Default |
|---|---|---|---|
| device | Yes | Device name from devices (op=list), e.g. dev_04f2 | |
| numbers | No | draw the element numbers on the screenshot | |
| with_text | No | also list the elements (numbered, centres in image pixels) |
App recipes: scripted flows that wait for the right screen at each step, deal with popups and report what they did. op=run runs a built-in recipe (package + action) or your own JSON and returns when it is finished. op=record_start begins recording what is done on the phone (through act, or by a person in the live view); record_status shows progress; record_stop returns the recorded draft recipe, ready for op=run.
| Name | Required | Description | Default |
|---|---|---|---|
| op | Yes | run | record_start | record_status | record_stop | |
| vars | No | run: variables as JSON, e.g. {"seconds": "300"} | |
| steps | No | run: your own recipe as JSON {app, steps:[...]} (e.g. a recorded draft) | |
| action | No | run: built-in recipe name (e.g. warm, login, publish_post) | |
| device | Yes | Device name from devices (op=list), e.g. dev_04f2 | |
| package | No | run: app package of a built-in recipe (e.g. com.instagram.android) |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Saved states of a phone, for starting every run from the same point. op=list: id, created, android, current, restorable. op=save: save the current state. op=restore with snapshot_id: reset the phone to that state (what happened since is lost). op=clone with snapshot_id and new_name: a new phone from that state.
| Name | Required | Description | Default |
|---|---|---|---|
| op | No | list (default) | save | restore | clone | list |
| device | Yes | Device name from devices (op=list), e.g. dev_04f2 | |
| new_name | No | clone: name of the new phone (letters and digits, up to 16) | |
| snapshot_id | No | restore / clone: id from op=list |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
act10 fields changedInput schema / properties / action / descriptionAdded value: +"one of: tap long_press swipe scroll tap_text wait_text type key open_app close_app open_url installed intent settings clipboard_set notifications_open notifications_close notifications_clear"Input schema / properties / device / descriptionAdded value: +"Device name from devices (op=list), e.g. dev_04f2"Input schema / properties / elementAdded value: +{
+ "anyOf": [
+ {
+ "type": "integer"
+ },
+ {
+ "type": "null"
+ }
+ ],
+ "default": null,
+ "description": "element number from the last observe (tap, long_press)",
+ "title": "Element"
+}Input schema / properties / ms / descriptionAdded value: +"swipe / long_press duration in ms"Input schema / properties / text / descriptionAdded value: +"type: the text; key: home|back|recents|enter|tab|del|menu; scroll: down|up; open_app/close_app: package; open_url: https://...; tap_text/wait_text: text on screen; settings: wifi|bluetooth|display|sound|apps|location|about|battery|...; intent: a URI to view (geo:, tel:, https:...) or JSON {action,data,package,component,extras}; clipboard_set: the text"Input schema / properties / timeout / descriptionAdded value: +"wait_text: seconds to wait, up to 120"Input schema / properties / x / descriptionAdded value: +"x in screenshot pixels (tap, long_press, swipe start)"Input schema / properties / x2 / descriptionAdded value: +"swipe end x"Input schema / properties / y / descriptionAdded value: +"y in screenshot pixels"Input schema / properties / y2 / descriptionAdded value: +"swipe end y"devices3 fields changedInput schema / properties / device / descriptionAdded value: +"device name, for start/stop"Input schema / properties / op / descriptionAdded value: +"list (default) | start | stop"Output schema / (root)Previous value: -nullNew value: +{
+ "properties": {
+ "result": {
+ "additionalProperties": true,
+ "title": "Result",
+ "type": "object"
+ }
+ },
+ "required": [
+ "result"
+ ],
+ "title": "devicesOutput",
+ "type": "object"
+}inspectobserve3 fields changedInput schema / properties / device / descriptionAdded value: +"Device name from devices (op=list), e.g. dev_04f2"Input schema / properties / numbersAdded value: +{
+ "default": true,
+ "description": "draw the element numbers on the screenshot",
+ "title": "Numbers",
+ "type": "boolean"
+}Input schema / properties / with_text / descriptionAdded value: +"also list the elements (numbered, centres in image pixels)"recipesnapshots5 fields changedInput schema / properties / device / descriptionAdded value: +"Device name from devices (op=list), e.g. dev_04f2"Input schema / properties / new_name / descriptionAdded value: +"clone: name of the new phone (letters and digits, up to 16)"Input schema / properties / op / descriptionAdded value: +"list (default) | save | restore | clone"Input schema / properties / snapshot_id / descriptionAdded value: +"restore / clone: id from op=list"Output schema / (root)Previous value: -nullNew value: +{
+ "properties": {
+ "result": {
+ "additionalProperties": true,
+ "title": "Result",
+ "type": "object"
+ }
+ },
+ "required": [
+ "result"
+ ],
+ "title": "snapshotsOutput",
+ "type": "object"
+}actdevicesobservesnapshotsCloud Android phones for AI agents: create a phone, read the screen, tap, type, install APKs, park.
Disposable cloud Android emulators for coding agents: run an APK or PR build, tap, type, screenshot.
Control real Android and iOS devices with LLM agents — tap, swipe, type, automate flows.
Melaya is a remote MCP server. It gives an assistant hands on your own Android phone and browser: it reads the screen through the accessibility tree, then taps, types and navigates inside the apps and sites you allow-list, with no per-app API. It also builds, schedules and runs agent pipelines across 6k+ connected tools. OAuth 2.1, nothing to install.
Add one secure layer between your agents and this server.
