# appdyno (npm · appdyno)

Test, drive and measure iOS and Android apps on simulators, emulators and real phones, from a Mac.

- Trust score: 59/100 (low)
- Change this week: −1
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-21

## Components

- npm · `appdyno`: 59/100 (this document), [markdown](https://verifymcp.io/servers/fatihcimen-appdyno/appdyno.md), [page](https://verifymcp.io/servers/fatihcimen-appdyno/appdyno)

## Channel facts

- Registry: `npm`
- Package: `appdyno`
- Version: `1.3.10`
- Transport: `stdio`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-11-21.

- **Supply Chain Security**: 50/100
  - Malware scan not yet available for this package.
  - No known CVEs affecting this package version or its production dependencies.
  - No install/post-install scripts declared.
  - No production dependencies, so there is no dependency health to assess.
- **Provenance & Transparency**: 100/100
  - Source repository is publicly reachable at the declared URL.
  - Cryptographically verified build provenance (signed, bound to fatihcimen/appdyno).
  - Clear OSI-approved license (MIT).
  - Actively maintained (last published 0 days ago).
  - Publishes a security disclosure policy (SECURITY.md).
- **Schema Quality & AI Usability**: 50/100
  - AI-judged instruction clarity (fair).
  - Context-footprint check failed: tool/resource definitions use about 5038 tokens (~179/item across 28 items; 28 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 0/100
  - Stability check failed: the tool surface changed between 1.0.0 and 1.3.10: 42 tool removals, 0 breaking changes, 22 additions.
- **Tool Coverage**: 70/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 10% of tool parameters carry a description.
- **Tool Safety**: 88/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - 1 of 2 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation; "transfer_files" implies "transfer" and declares no destructiveHint at all, which the MCP spec reads as destructive by default.
  - An AI judge read all 29 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a current MCP spec version (2026-07-28).

## Install

### How do I install the appdyno MCP server?

appdyno runs locally as an npm package, launched with npx -y appdyno mcp. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add fatihcimen-appdyno -- npx -y appdyno mcp
```

### Cursor

```json
{
  "mcpServers": {
    "fatihcimen-appdyno": {
      "command": "npx",
      "args": [
        "-y",
        "appdyno",
        "mcp"
      ]
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "fatihcimen-appdyno": {
      "command": "npx",
      "args": [
        "-y",
        "appdyno",
        "mcp"
      ]
    }
  }
}
```

### Codex

```bash
codex mcp add fatihcimen-appdyno -- npx -y appdyno mcp
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "fatihcimen-appdyno": {
      "type": "local",
      "command": [
        "npx",
        "-y",
        "appdyno",
        "mcp"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add fatihcimen-appdyno --command npx --arg -y --arg appdyno --arg mcp
```

### Hermes

```yaml
mcp_servers:
  fatihcimen-appdyno:
    command: "npx"
    args: ["-y", "appdyno", "mcp"]
```

### Netclaw

```json
{
  "McpServers": {
    "fatihcimen-appdyno": {
      "Transport": "stdio",
      "Command": "npx",
      "Arguments": [
        "-y",
        "appdyno",
        "mcp"
      ]
    }
  }
}
```

### Vellum

```bash
assistant mcp add fatihcimen-appdyno -t stdio -c npx -a -y appdyno mcp
```

### Other

```json
{
  "mcpServers": {
    "fatihcimen-appdyno": {
      "command": "npx",
      "args": [
        "-y",
        "appdyno",
        "mcp"
      ]
    }
  }
}
```

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-11-21 (score 59, +11)

- [security improvement] Malware scan: unverified → pass
- [security improvement] Known CVEs: unverified → pass
- [functional improvement] Dependency health: unverified → pass
- [functional] Package version: 1.3.8 → 1.3.10

### 2026-11-20 (score 48, −15)

- [security regression] Malware scan: pass → unverified
- [security improvement] Known CVEs: unverified → pass
- [functional improvement] Dependency health: unverified → pass
- [functional] Package version: 1.3.6 → 1.3.8
- [functional] Package version: 1.3.6 → 1.3.7

### 2026-11-19 (score 63, +4)

- [security regression] Known CVEs: pass → unverified
- [security regression] Stability: fail → unverified
- [security regression] Tool safety: pass → unverified
- [security improvement] Malware scan: unverified → pass
- [functional regression] Dependency health: pass → unverified
- [functional regression] Capabilities: pass → unverified
- [functional regression] Tool coverage: 100 → unverified
- [functional] First check of Schema quality: unverified
- [functional] Package version: 1.3.2 → 1.3.6
- [functional] Package version: 1.3.2 → 1.3.5
- [functional] Package version: 1.3.2 → 1.3.3

### 2026-11-18 (score 59, 0)

- [security improvement] Malware scan: unverified → pass
- [functional] Package version: 1.3.0 → 1.3.2

### 2026-11-17 (score 59, −17)

- [security regression] Malware scan: pass → unverified
- [security regression] Stability: 0.13 → fail
- [functional regression] Schema quality: 103 → 179
- [functional improvement] Tool coverage: 4% → 10%
- [functional improvement] Destructive annotations: 0% → 50%
- [functional] Package version: 1.2.0 → 1.3.0

### 2026-11-16 (score 76, +11)

- [security regression] Tool safety: pass → unverified
- [security regression] Stability: 0.10 → unverified
- [security improvement] Known CVEs: unverified → pass
- [functional regression] Capabilities: pass → unverified
- [functional regression] Tool coverage: 100 → unverified
- [functional regression] Schema quality: pass → fail
- [functional improvement] Dependency health: unverified → pass
- [functional] First check of Schema quality: unverified
- [functional] Package version: 1.1.3 → 1.2.0

### 2026-11-15 (score 65, +5)

- [security regression] Known CVEs: pass → unverified
- [security improvement] Malware scan: unverified → pass
- [functional regression] Dependency health: pass → unverified
- [functional] Package version: 1.1.1 → 1.1.3
- [functional] Package version: 1.1.1 → 1.1.2

### 2026-11-14 (score 60, −14)

- [security regression] Malware scan: pass → unverified
- [security regression] Known CVEs: pass → unverified
- [functional regression] Dependency health: pass → unverified
- [functional improvement] Stability: unverified → 0.07
- [functional] Package version: 1.0.0 → 1.1.1
- [functional] Package version: 1.0.0 → 1.1.0

## MCP tools (28)

### `list_devices` (~81 tokens)

List devices

List connected phones, simulators and emulators. Returns each one's serial, state, model, whether this server may drive it, who holds it, what blocks it and the fix. selected: only the device in use, with OS, screen, agent version and the app in front. One not running: manage_simulator.

Input parameters:

- `selected` (boolean)

### `use_device` (~96 tokens)

Use device

Select the device later calls drive, by a serial list_devices gives; app: the app id under test. Not needed with one device connected. One server drives one device: for several in parallel, a server each. A device another process holds is refused, naming it; share: true drives it anyway. Returns what is selected.

Input parameters:

- `app` (string)
- `device` (string, required)
- `share` (boolean)

### `manage_simulator` (~88 tokens)

Simulators and emulators

Start or stop an iOS simulator or Android emulator, only on request. list: what can be started; boot (name: id or name); create (name: an iOS model; boots and selects it); shutdown (name; none: those this tool started). Connected devices: list_devices. Returns the device.

Input parameters:

- `action` (string, required)
- `name` (string)

### `set_device` (~261 tokens)

Device conditions

Read or set a device condition: dark mode, orientation (landscape), GPS location, clipboard, Face ID, fingerprint, memory warning, shake, text size, locale, slow network, airplane mode, Wi-Fi, battery, fold, calls, SMS. No value reads it. App permissions: manage_app; photos: transfer_files. Returns the value; orientation the screen too. Put back at the session end or by restore, but for the clipboard.

Input parameters:

- `app` (string)
- `name` (string, required)
- `read` (string)
- `title` (string)
- `value` (string): orientation: portrait, landscape_left, upside_down, landscape_right; location: lat,lon, none; clipboard: its text. Switch: on, off; text_size: Android 1.3, iOS extra-large; locale: tr-TR; network: sl…

### `read_screen` (~119 tokens)

Read screen

Read the screen as text. Returns each element of the accessibility tree as ref role "label" x,y flags; views: the native view hierarchy instead. selector: only its matches, with state, value and bounds. Assert: check_element; pixels: take_screenshot; audits, visual diffs: check_screen; web views: inspect_app. Like rows fold under a header; refs hold across scrolls. raw: every node as data.

Input parameters:

- `raw` (boolean)
- `selector` (object)
- `views` (boolean)

### `take_screenshot` (~171 tokens)

Screenshot

An image of the screen, for a visual judgement only: read_screen is cheaper and gives refs. Returns it (at most 800 px unless width) with its scale to tap units; points: 1 pixel is 1 unit. watchMs: no image, what moved (way, speed, jumps); strip: 8 timed frames. file: full size to a .png, .jpg or folder; inline: false hides it; overwrite. An unchanged screen gets no second image. Visual diffs: check_screen; video: record_session.

Input parameters:

- `file` (string)
- `inline` (boolean)
- `overwrite` (boolean)
- `points` (boolean)
- `strip` (boolean)
- `watchMs` (integer)
- `width` (integer)

### `check_element` (~132 tokens)

Check or wait for an element

Assert or wait for an element, touching nothing: check exists, absent, text (equals), enabled, checked, focused, moves; expect: false for the opposite. timeoutMs (at most 15000) waits until it holds. Returns one line, or fails with CHECK_FAILED saying what it found. Whole screen: check_screen; look only: read_screen.

Input parameters:

- `check` (string)
- `equals` (string)
- `expect` (boolean)
- `read` (string)
- `ref` (string)
- `selector` (object)
- `timeoutMs` (integer)

### `check_screen` (~142 tokens)

Check the whole screen

Check the whole screen. accessibility: an accessibility audit of contrast, touch targets, labels, reading order (types, maxIssues). visual: a visual regression test against the image name in dir, failing with CHECK_FAILED and a diff image past tolerance (percent of pixels); masks are left out. visual_baseline saves it. One element: check_element; pixels: take_screenshot. Returns the issues or the diff.

Input parameters:

- `check` (string, required)
- `dir` (string)
- `masks` (array)
- `maxIssues` (integer)
- `name` (string)
- `tolerance` (number)
- `types` (array)

### `tap_screen` (~125 tokens)

Tap

Tap a target in place: ref, selector or x,y. count: 2 is a double tap; holdMs: a long press. A moving finger: swipe_screen; a system alert: answer_alert. Returns the screen after. Several matches refuse, listing refs; no match waits briefly, then refuses with the nearest labels.

Input parameters:

- `count` (number)
- `holdMs` (integer)
- `read` (string)
- `ref` (string)
- `selector` (object)
- `x` (number)
- `y` (number)

### `swipe_screen` (~233 tokens)

Swipe, scroll, drag, pinch

Move a finger: swipe, scroll to an element, drag or pinch. direction: a swipe, from a target if given; up moves content up. until: swipes until that selector is on screen (maxSwipes, 10); x,y: which list. to: drags the target there, no flick; holdMs first lifts a row. scale: pinch, > 1 zooms in. pointers: free paths, one per finger, t in ms. exact: content moves distance and stops. In place: tap_screen. Returns the screen after.

Input parameters:

- `direction` (string)
- `distance` (number)
- `durationMs` (integer)
- `exact` (boolean)
- `holdMs` (integer)
- `maxSwipes` (integer)
- `pointers` (array)
- `read` (string)
- `ref` (string)
- `scale` (number)
- `selector` (object)
- `to` (object)
- `until` (object)
- `x` (number)
- `y` (number)

### `type_text` (~111 tokens)

Type text

Type up to 500 characters of Unicode into the focused field, or a target, tapped and focused first. replace: the field's text; submit: then enter. One key or button: press_key. Returns the screen after.

Input parameters:

- `read` (string)
- `ref` (string)
- `replace` (boolean)
- `selector` (object)
- `submit` (boolean)
- `text` (string, required)
- `x` (number)
- `y` (number)

### `press_key` (~119 tokens)

Press key

Press a key or button: back, home, enter, delete, tab, escape, space, arrow_up/down/left/right, app_switch, volume_up/down, mute, siri, power, lock; a letter or digit, held keys first: command+a (Android: Ctrl). hide_keyboard closes the on-screen keyboard. control_center, notification_center, spotlight open those; home closes them. Words: type_text. Returns the screen after.

Input parameters:

- `count` (integer)
- `key` (string, required)
- `read` (string)

### `run_steps` (~110 tokens)

Run steps

Run known actions as one call: at most 30, each { tool: a tool's name, ...its arguments }, as { tool: tap_screen, selector: { text: OK } }; any tool but run_steps and run_flow. Stops at the first failure, saying which and why. Returns every answer, images too; a step reads the screen only with read on it, or last. As a test: run_flow.

Input parameters:

- `read` (string)
- `steps` (array, required)

### `answer_alert` (~89 tokens)

System alert

Read or answer the open system alert, as a permission prompt; an app's own dialog: tap_screen. answer: a button's label to touch, or allow (least: once, else while using) or deny, in any language. Returns its text and buttons, or the screen after; NOT_FOUND if none is open.

Input parameters:

- `answer` (string)
- `read` (string)

### `ask_person` (~95 tokens)

Ask the person

Ask your person to do what no tool can (Face ID, a passcode) and wait until they say done or, with no window, the screen changes. A system alert: answer_alert. On this Mac: a banner and a sound in the live window, else a sound and a notification of reason. Returns when done; refuses at timeoutS.

Input parameters:

- `reason` (string, required)
- `timeoutS` (integer)

### `show_mirror` (~85 tokens)

Live window

Open a live screen mirror for your person, who can touch it as you drive. open, close; status: what stands in its way, the fix; stats: true frame rates. It gives you no frames: take_screenshot, record_session. Returns its state.

Input parameters:

- `action` (string, required)
- `quality` (string)
- `wifi` (boolean)

### `launch_app` (~140 tokens)

Launch app

Launch an app by its id or the name manage_app list gives, to the front on the screen it was on if it runs. restart: start it fresh, as args (Android: key=value extras), env and language (tr, pt-BR) do. probe: with the view probe read_screen views reads. By URL: open_link; to stop it: manage_app. Returns the screen after.

Input parameters:

- `app` (string)
- `args` (array)
- `env` (object)
- `language` (string)
- `probe` (boolean)
- `read` (string)
- `restart` (boolean)

### `manage_app` (~213 tokens)

Manage apps

Install and manage apps: list (all: system apps too), state, install (path: .apk, .app, .ipa; files: an .apk with its .dm or split APKs), uninstall, stop (background: as the system ends it, state kept), clear (data), grant, revoke (service: location, photos…; a wrong word lists them), build_and_run (path: Xcode project or Gradle folder; scheme or module; configuration or variant; args, env). Starting one: launch_app; its data: transfer_files. Returns what was done.

Input parameters:

- `action` (string, required)
- `all` (boolean)
- `app` (string)
- `args` (array)
- `background` (boolean)
- `configuration` (string)
- `env` (object)
- `files` (array)
- `path` (string)
- `read` (string)
- `scheme` (string)
- `service` (string)

### `transfer_files` (~110 tokens)

Transfer files

File transfer between this Mac and the device. pull: download an app's data to the folder path. push: upload the folder path as its data, replacing it (stops the app; asks the person). media: add the photos or videos at path to the gallery, taken out at the session end. An .apk or .ipa: manage_app install. Returns what moved, and where.

Input parameters:

- `action` (string, required)
- `app` (string)
- `path` (string, required)

### `open_link` (~116 tokens)

Open link

Open one of: url, a web URL, deep link, universal link or App Link, in the selected app if any (unknown schemes ask first); settings, a Settings page (about, privacy) or an app's (its id, or app); links, an app's id: the links it opens. An app by id: launch_app. Returns the screen after, or links.

Input parameters:

- `links` (string)
- `read` (string)
- `settings` (string)
- `url` (string)

### `send_push` (~140 tokens)

Send push notification

Send the app a push notification: remote notification, rich notification, Live Activity, by APNs or FCM. No type: lists what this device takes and how, sending nothing. title, body, image, data; or payload, APNs' or FCM's own JSON. Returns what was sent, and the screen after.

Input parameters:

- `app` (string)
- `body` (string)
- `data` (object)
- `image` (string)
- `payload` (string): APNs' or FCM's JSON
- `read` (string)
- `title` (string)
- `type` (string)

### `read_logs` (~169 tokens)

Logs and crashes

Read the app's log, or with source crashes its crash, ANR and exit reports. Returns, for the log, its own lines (logcat on Android), newest last, secrets redacted; level, contains, limit (200). For crashes: a bounded summary and each report's file; limit (5), symbols. An action the app died under already says why: this is for more. Requests: capture_network.

Input parameters:

- `app` (string)
- `contains` (string)
- `level` (string)
- `limit` (integer)
- `since` (string): ISO time; default: since the last action
- `source` (string)
- `symbols` (string): a .dSYM or folder of them; Android: a folder of .so

### `capture_network` (~472 tokens)

Network traffic

Capture an app's network traffic (HTTP), mock its answers, throttle it. paths: what each way sees here, or why not; start (via proxy: at proxy host:port; probe: restarts the app; bodies; strict); read (id: one request whole; contains, method, status; har: a proxy's export; save: a .har); check: waits for a request, failing with the nearest; mock (rules; add; from: a file of answers; unmatched); throttle; stop. The whole device: set_device network. Returns requests, bounded; installs no certificate.

Input parameters:

- `action` (string, required)
- `add` (boolean)
- `after` (integer): only requests after this id
- `app` (string)
- `bodies` (boolean)
- `contains` (string)
- `diff` (integer): against this id
- `from` (string): rules from a request id or a .har
- `har` (string)
- `has` (string): in its body: path=value or text
- `id` (integer)
- `limit` (integer)
- `method` (string)
- `overwrite` (boolean)
- `path` (string): a body value: data.items[0].id
- `proxy` (string)
- `rules` (array): Each "host/path* [method=] [query=] [regex=] [has=] [when=]" then "status= body= (@file) type= header=", "fail=offline|timeout|reset" or "to=host", and delayMs= times= off. Replaces; [] clears; none…
- `save` (string)
- `shape` (boolean): keys and types, no values
- `status` (string): 404, 4xx, failed
- `strict` (boolean): a 5xx or a failed request fails the action
- `throttle` (string): full, slow, edge, 3g, lte, offline, bytes/s down, or "delayMs= down= up= loss="
- `timeoutMs` (integer)
- `unmatched` (string): a request no rule takes
- `via` (string)

### `inspect_app` (~181 tokens)

Inside an app

Reach inside an app where read_screen cannot: a web view, React Native, or an iPhone with no runner. Web view of a debuggable app: web_pages, web_elements (the DOM, at tap coordinates), web_click (selector: CSS), web_eval (js); page: id or part of a title or URL. Metro: rn_targets, rn_logs (the JS console for ms), rn_reload. The iPhone's accessibility service: ax_list, ax_press (name). Returns the listing, or what was done.

Input parameters:

- `action` (string, required)
- `app` (string)
- `js` (string)
- `metro` (string)
- `ms` (integer)
- `name` (string)
- `page` (string)
- `selector` (string)
- `target` (string)

### `record_session` (~148 tokens)

Video and report

Keep evidence. kind video: screen video, one per device; start, stop. report: a test report, index.html with steps, video, logs, crashes, measurements, secrets hidden; start, write, stop. timeline: start, the journey, stop: timeline.html, with video, touches, frames, requests, log and memory on one clock. flight: the last seconds, kept on a crash or stall. To replay actions: run_flow. Returns the file.

Input parameters:

- `action` (string, required)
- `dir` (string)
- `file` (string)
- `kind` (string, required)
- `maxSeconds` (integer)
- `overwrite` (boolean)

### `measure_performance` (~363 tokens)

Measure performance

Measure performance by metric. launch: startup time, cold, warm or hot, over repeats; trace: why. scroll: jank and frozen frames of a list (x,y or selector). heap: memory leaks, the heap to a file. frames: why frames were slow, hitches, hangs (what). cpu: top functions over seconds. snapshot: CPU, memory, frames, load and heat now (also). Returns numbers, findings, a trace's path; scenario keeps the run for compare_performance.

Input parameters:

- `also` (array): energy (says if a model); anonMemory: by process state, bitmaps; surface: a SurfaceView or game; startInfo: why it last started; battery: wake locks, wakeups, jobs
- `app` (string)
- `baseline` (boolean): 3 runs, added to compare_performance's baseline
- `baselineProfile` (boolean): Android, a release build: in the APK, applied, cold starts with and without
- `batteryWindowMs` (integer)
- `compile` (string)
- `cpuWindowMs` (integer)
- `direction` (string): frames, cpu: scroll meanwhile
- `dsym` (string): the build's .dSYM or a folder of them
- `metric` (string, required)
- `mode` (string)
- `probe` (boolean)
- `profile` (string)
- `repeats` (integer)
- `scenario` (string)
- `seconds` (number)
- `selector` (object)
- `swipes` (integer)
- `trace` (boolean)
- `what` (string)
- `x` (number)
- `y` (number)

### `compare_performance` (~259 tokens)

Performance baseline and gate

Performance regression gate on runs measure_performance kept. compare (default): the last runs against baseline and budgets in .appdyno/perf/, failing a change statistically and practically significant. baseline: keeps the last runs (3 needed; replace anew); budget: a kind's max, min. why: a regression's cause by its flow, restarting the app; reference: its baseline side. functions: two cpu ids, from, to. query: a bounded table of a kept trace (read-only sql, table). prune: traces. Returns the verdict.

Input parameters:

- `action` (string)
- `app` (string)
- `deviceClass` (string)
- `dryRun` (boolean)
- `dsym` (string)
- `format` (string)
- `from` (string)
- `kind` (string)
- `last` (integer)
- `limit` (integer)
- `limitGb` (number)
- `max` (number)
- `min` (number)
- `replace` (boolean)
- `scenario` (string)
- `sql` (string)
- `table` (string)
- `to` (string)
- `trace` (string)

### `run_flow` (~188 tokens)

Record and replay a flow

Record your calls as a repeatable test, a .flow file replayed with no model: record, then stop to write it to file (save: later); run replays file (from: a step; iterations: per step for compare_performance); test: a file or folder to one verdict, report in to; export: Maestro YAML. params NAME=value: recorded, a value becomes ${NAME}; run fills it. One-off calls: run_steps. Returns the steps, times or verdict.

Input parameters:

- `action` (string, required)
- `envFile` (string)
- `file` (string)
- `from` (number)
- `iterations` (integer)
- `overwrite` (boolean)
- `params` (array)
- `retries` (integer)
- `scenario` (string)
- `tags` (array)
- `to` (string)

## Diagnostics

Captured diagnostic sections: Provenance, Dependencies. The full working is on the page: https://verifymcp.io/servers/fatihcimen-appdyno/appdyno#diagnostics

## Score history

- 2026-11-21: 59
- 2026-11-20: 48
- 2026-11-19: 63
- 2026-11-18: 59
- 2026-11-17: 59
- 2026-11-16: 76
- 2026-11-15: 65
- 2026-11-14: 60
- 2026-11-13: 74
- 2026-11-12: 59

## Common questions

### What is the appdyno MCP server?

appdyno is an MCP server listed in the public MCP registry as io.github.fatihcimen/appdyno. Test, drive and measure iOS and Android apps on simulators, emulators and real phones, from a Mac. This page covers its npm package (appdyno).

### Is the appdyno MCP server safe to use?

appdyno scores 59 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 21 November 2026. It declares no install or post-install scripts. Its build provenance is signed and verified. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the appdyno MCP server expose?

appdyno exposes 28 tools: list_devices, use_device, manage_simulator, set_device, read_screen, and 23 more. Their descriptions and schemas cost roughly 4,556 tokens of context every time the server is loaded.

### Is the appdyno MCP server still maintained?

appdyno is still listed as active in the MCP registry. We last reached this channel on 21 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What licence is the appdyno MCP server under?

appdyno declares the MIT licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.

### What did VerifyMCP test on the appdyno MCP server?

VerifyMCP ran 15 checks on appdyno, covering Known CVEs, Install scripts, Dependency health, Source repository and 11 more. One item could not be completed, a gap on our side rather than a finding about appdyno: Malware scan. Endpoint Security and Transport & Reachability do not apply to an npm package, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.

## Links

- npm package: https://www.npmjs.com/package/appdyno
- Socket report: https://socket.dev/npm/package/appdyno
- Repository: https://github.com/fatihcimen/appdyno
- Changelog RSS feed: https://verifymcp.io/servers/fatihcimen-appdyno/appdyno.xml
- Changelog JSON feed: https://verifymcp.io/servers/fatihcimen-appdyno/appdyno/changelog.json
- Trust report (JSON): https://verifymcp.io/servers/fatihcimen-appdyno/appdyno.json
- HTML version of this page: https://verifymcp.io/servers/fatihcimen-appdyno/appdyno
