Un servidor MCP para agentes de codificación de IA (Claude Code, Cursor, etc.) que permite probar aplicaciones Android cuya interfaz de usuario completa se dibuja dentro de una única superficie opaca OpenGL/Vulkan/Metal — Cocos2d-x, Unity, Unreal, OpenGL puro, libGDX y motores similares.
adb shell uiautomator dump y todas las herramientas de automatización basadas en el árbol de accesibilidad (incluida la mayoría de los servidores MCP de automatización móvil) funcionan inspeccionando la jerarquía de vistas nativa de Android: botones, etiquetas, su texto y coordenadas. Eso funciona muy bien para una interfaz Android normal construida con vistas nativas.
No funciona para un juego o aplicación que renderiza toda su interfaz como texturas dentro de un único GLSurfaceView. Desde el punto de vista del árbol de accesibilidad, hay exactamente una vista opaca en pantalla sin hijos, sin etiquetas, sin coordenadas para nada dentro de ella. No hay nada que inspeccionar: la pantalla es una caja negra, sin importar cuánta interfaz haya realmente en ella.
El único canal de observación real que queda son las capturas de pantalla. Este servidor está construido en torno a ese hecho como el caso normal, no como un recurso ocasional.
mobile-next/mobile-mcp es el servidor MCP de automatización móvil de propósito general, y es una buena opción por defecto para aplicaciones nativas normales: primero el árbol de accesibilidad (rápido, barato, sin modelo de visión, sin tokens de imagen), con respaldo a capturas de pantalla + coordenadas solo cuando el árbol no le da lo que necesita.
Para una aplicación con lienzo OpenGL, ese respaldo no es ocasional: es la única ruta que funciona, cada vez. mobile-mcp-opengl está construido específicamente para ese caso y, como resultado, toma dos decisiones de diseño diferentes:
Sin intento de árbol de accesibilidad en absoluto. No hay nada que ganar intentándolo: siempre vuelve vacío para estas aplicaciones, así que cada herramienta aquí va directamente a captura de pantalla + visión.
El análisis de visión pasa por un proveedor separado y conectable (ver más abajo), no por el modelo que ejecuta el agente llamante. Un bucle de control de calidad funcional sobre un juego puede fácilmente llegar a cientos de comprobaciones de capturas de pantalla por sesión; enrutar todo eso a través de la propia visión de tu agente de codificación principal cuesta tanto dinero real como tokens/contexto que preferirías gastar en el trabajo de codificación real. Aquí los bytes de la captura de pantalla nunca entran en el contexto del agente llamante; solo lo hace la respuesta de texto corta del proveedor.
Related MCP server: Android-MCP
Un diseño ingenuo expone tap, screenshot y ask como tres herramientas separadas. Eso obliga al agente llamante a orquestar un bucle de varios pasos para cada interacción: tocar → tomar una captura de pantalla → pasarla a un paso de visión → leer el resultado → decidir qué hacer a continuación. Cada uno de esos es una llamada de herramienta separada y un turno separado: quemando tokens en coordinación en lugar de en la lógica de prueba real, y dando más superficie para que el agente omita un paso, los desordene o razone sobre estado obsoleto entre llamadas.
En cambio, este servidor expone herramientas combinadas — tap_and_ask, swipe_and_ask, long_press_and_ask — que realizan la acción, esperan brevemente, toman la captura de pantalla, preguntan al proveedor de visión y devuelven una respuesta corta, todo como una sola llamada de herramienta. Un escenario de prueba de varios pasos termina costando aproximadamente un turno de agente por comprobación significativa, no tres o cuatro.
También están disponibles screenshot_ask simple (solo observación, sin acción) y herramientas baratas sin visión (type_text, press_key, logcat_grep) para las partes de un flujo de prueba que no necesitan este patrón.
Herramienta | Qué hace | ¿Llamada de visión? |
| Captura de pantalla, luego pregunta una pregunta corta sobre ella | Sí |
| Toca (x, y), espera, captura, pregunta | Sí |
| Desliza/arrastra (x1,y1)→(x2,y2), espera, captura, pregunta | Sí |
| Mantén presionado (x, y) durante una duración, espera, captura, pregunta | Sí |
| Acción opcional, luego N capturas de pantalla espaciadas en el tiempo, pregunta lo mismo sobre cada fotograma | Sí (N llamadas) |
| Escribe en el campo actualmente enfocado | No |
| Envía un evento | No |
| Lee logcat reciente, opcionalmente filtrado por regex | No |
| Informa el gasto acumulado de visión de hoy y los umbrales | No |
Prefiere logcat_grep sobre una llamada de visión siempre que lo que necesitas ya esté en una línea de registro (errores, tus propias impresiones de depuración, errores de red) — es gratis y exacto, una llamada de visión no es ninguna de las dos cosas.
record_and_askLas herramientas de un solo fotograma no pueden decirte si algo anima correctamente (¿el indicador de fuerza pulsa suavemente, una etiqueta vuela hacia arriba y se desvanece, un sprite vuelve a su posición inicial?). record_and_ask realiza una acción opcional (tocar o deslizar, o ninguna), espera waitMs (mismo significado que waitMs en tap_and_ask/swipe_and_ask — tiempo para que la interfaz comience a reaccionar antes del primer fotograma), luego captura frameCount capturas de pantalla espaciadas intervalMs aparte, y devuelve una respuesta corta por fotograma — el agente llamante obtiene una línea de tiempo en una sola llamada de herramienta en lugar de orquestar N viajes de ida y vuelta de captura de pantalla + pregunta por sí mismo.
Por qué una llamada de visión por fotograma, no una llamada con todos los fotogramas juntos. Resulta que imageCaption de Runware acepta un array inputImages (plural) no documentado junto con el inputImage único documentado — probado directamente contra la API. Funciona limpiamente para exactamente 2 imágenes (una comparación antes/después en la misma solicitud volvió correcta y coherente). Con 3+ imágenes en una solicitud, tanto ese parámetro de array como una imagen de "tira de película" compuesta manualmente lado a lado produjeron respuestas truncadas o malformadas en las pruebas — el pequeño modelo de visión de 7B aparentemente pierde coherencia más allá de cierta carga combinada de visual+instrucción en una llamada. Las llamadas secuenciales de imagen única (el enfoque de esta herramienta) fueron confiables en cualquier número de fotogramas probado, y no son significativamente más caras: el costo está dominado por la longitud de la respuesta (ver más abajo) en lugar del número de llamadas, así que N respuestas cortas secuenciales cuestan aproximadamente lo mismo, o menos, que una respuesta larga de múltiples imágenes. Si tu propio proveedor maneja solicitudes de múltiples imágenes de manera más confiable, este es un lugar obvio para optimizar — ver "Trae tu propio modelo".
git clone <this repo>
cd mobile-mcp-opengl
npm install
cp .env.example .env
# edit .env: at minimum set RUNWARE_API_KEY (or switch VISION_PROVIDER, see below)Requiere adb en PATH (o ADB_PATH configurado en .env), y un dispositivo o emulador en ejecución/conectado. Si hay más de uno conectado, configura ADB_DEVICE_SERIAL (ver adb devices).
Añade un .mcp.json en la raíz de tu proyecto (este archivo suele ser local al proyecto y estar ignorado por git, ya que normalmente apunta a una ruta específica de la máquina o contiene anulaciones de entorno específicas de la máquina):
{
"mcpServers": {
"mobile-opengl": {
"command": "node",
"args": ["/absolute/path/to/mobile-mcp-opengl/src/server.js"]
}
}
}Claude Code lo recoge automáticamente para el proyecto. El servidor lee su propio .env (junto a package.json en este repositorio) para toda la configuración — el agente llamante nunca necesita conocer ni pasar ninguna clave de API por sí mismo.
La longitud de la respuesta impulsa el costo, no el tamaño de la imagen. Esto se midió empíricamente contra el proveedor predeterminado Runware/Qwen2.5-VL-7B-Instruct: la misma pregunta con una respuesta forzada de una palabra costó lo mismo ($0.0006) en tamaños de imagen desde 360×360 hasta 1600×2400 (clase retina). La misma imagen de 1024×1024 con un prompt abierto de "describe esto" costó $0.0013–0.0019 — 2-3 veces más — puramente porque el modelo escribió una respuesta más larga, no porque la imagen fuera más grande.
Implicaciones prácticas:
No te molestes en reducir la resolución de las capturas de pantalla antes de enviarlas — no reduce significativamente el costo para este proveedor, y pierdes detalle que podrías necesitar.
Siempre formula preguntas para forzar respuestas cortas: sí/no, un número, una etiqueta corta, un pequeño objeto JSON con un par de campos. Cada herramienta en este servidor añade automáticamente una instrucción de respuesta corta, pero una pregunta vaga y abierta ("¿qué ves?") aún puede empujar al modelo hacia una respuesta más larga que una específica ("¿es visible el diálogo de error? sí/no").
A ~$0.0006/llamada para preguntas cortas bien formadas, una sesión de control de calidad de 500 llamadas cuesta aproximadamente $0.30. El mismo volumen de preguntas abiertas de "describe la pantalla" puede costar 2-3 veces más.
Cada llamada de visión se registra en .vision-log.jsonl (JSONL, una entrada por llamada: marca de tiempo, pregunta, respuesta, costo). Dos protecciones independientes se asientan sobre ese registro, ambas independientes del proveedor (funcionan con lo que sea que un proveedor reporte como costUsd):
Alerta por llamada (VISION_ALERT_USD, por defecto $0.0015): si una sola llamada vuelve por encima de esto, la respuesta de la herramienta incluye una nota [COST ALERT] que te dice que el modelo probablemente ignoró la instrucción de respuesta corta — una señal para reformular la pregunta, no algo que tragarse en silencio.
Tope diario (VISION_SESSION_CAP_USD, por defecto $2.00): una vez que el gasto acumulado de hoy alcanza esto, cada llamada de visión adicional es rechazada directamente (antes de llegar al proveedor) hasta que se aumente el tope o cambie el día. Esto es un freno duro contra un bucle descontrolado, no solo una advertencia.
Llama a vision_spend_report en cualquier momento para verificar el total de hoy sin hacer una llamada de dispositivo o visión.
Si un proveedor no puede reportar un costo (ver openai-compatible más abajo), las llamadas de él se registran con costUsd: null y nunca activan la alerta ni cuentan para el tope — las salvaguardas simplemente no pueden proteger un gasto del que no tienen visibilidad.
El análisis de visión pasa por src/providers/visionProvider.js, que elige un proveedor por nombre desde VISION_PROVIDER en .env. Hay dos integrados:
runware (predeterminado) — habla directamente con la tarea imageCaption de Runware.ai, usando Qwen2.5-VL-7B-Instruct (ID AIR runware:152@2) por defecto. Runware y OpenRouter son dos servicios separados con claves de API y catálogos de modelos separados — esto habla con Runware directamente, no a través de OpenRouter.
openai-compatible — un proveedor genérico para cualquier cosa que hable el formato de visión de chat-completions de OpenAI (partes de contenido image\_url). Funciona con OpenRouter, un servidor local de Ollama/LM Studio que ejecuta un modelo de visión, Groq, Together.ai o cualquier otro endpoint compatible. Configura OPENAI_COMPATIBLE_BASE_URL, OPENAI_COMPATIBLE_API_KEY, OPENAI_COMPATIBLE_MODEL en .env. La mayoría de las APIs compatibles con OpenAI reportan uso de tokens en lugar de un costo en dólares plano; configura OPENAI_COMPATIBLE_PRICE_PER_1M_INPUT/_OUTPUT si quieres que este proveedor estime costUsd a partir de eso (de lo contrario, el seguimiento de costos/las salvaguardas son inertes para este proveedor, según la nota anterior).
Para añadir un proveedor completamente personalizado (un modelo autoalojado, una forma de API completamente diferente), copia src/providers/openaiCompatibleProvider.js como punto de partida, implementa:
async function ask(imageBuffer, mimeType, question) {
// return { text: string, costUsd: number | null }
}
module.exports = { ask };y regístralo con un nombre en loadProvider() de src/providers/visionProvider.js.
MIT
Desarrollado por Kinect.PRO
Read the last N logcat lines, optionally filtered by a regex (e.g. your app's tag, or "Exception|FATAL"). No vision call, no cost - prefer this over screenshot_ask whenever what you need is already in a log line (crashes, your own debug prints, network errors).
| Name | Required | Description | Default |
|---|---|---|---|
| lines | No | How many recent lines to fetch (default 200). | |
| filterRegex | No | Optional regex; only matching lines are returned. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden. It clearly frames the operation as a read ('Read the last N logcat lines'), implying no mutation, and adds resource-related behavior ('No vision call, no cost'). It does not detail empty-result behavior or regex error handling, but for a non-destructive log reader this is sufficient.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two compact sentences, each earning its place: the first states the core operation, the second provides selection guidance and cost context. No redundant phrases or unnecessary details.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given only two optional parameters and no output schema, the description covers the action, filtering, selection criteria, and cost trade-off. It is complete enough for an agent to invoke correctly, though it does not spell out behavior for empty results or invalid regex.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema already documents both parameters with 100% coverage, so the baseline is 3. The description adds practical regex examples ('your app's tag, Exception|FATAL') and clarifies that the filter is optional, providing contextual guidance beyond the raw schema.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description uses a specific verb ('Read') and a clear resource ('logcat lines'), and it explicitly distinguishes itself from a sibling tool (screenshot_ask) by stating 'No vision call, no cost'. An agent can immediately tell what this tool does and how it differs.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
It gives an explicit when-to-use rule: 'prefer this over screenshot_ask whenever what you need is already in a log line,' followed by concrete examples (crashes, debug prints, network errors). It also explains the cost advantage, making the selection decision clear.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Long-press at (x, y) for durationMs, wait briefly, take a screenshot, and ask a short question about the result.
| Name | Required | Description | Default |
|---|---|---|---|
| x | Yes | ||
| y | Yes | ||
| waitMs | No | Milliseconds to wait after releasing before screenshotting (default 500). | |
| question | Yes | ||
| durationMs | No | Hold duration in ms (default 800). |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the burden of disclosing behavior. It transparently lists the operation sequence and references default waitMs/durationMs defaults in the schema. However, it does not disclose side effects of long-pressing (e.g., opening context menus or triggering navigation), what the 'ask' returns or to whom, or any required permissions.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
A single sentence that front-loads the core action and includes the full workflow without filler. Every element earns its place.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
With 5 parameters, no annotations, and no output schema, the description leaves important gaps: the return/response behavior is ambiguous ('ask a short question about the result'), coordinate system is unspecified, and side effects are not mentioned. An agent would need additional implicit knowledge to call this tool confidently.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema descriptions only cover waitMs and durationMs (40% coverage). The description helps by framing x and y as long-press coordinates and question as a short question about the result. Still, it does not specify coordinate units/origin or any constraints on the question, so it only partially compensates for the schema gaps.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States a specific action sequence: long-press at (x, y) for durationMs, wait, screenshot, and ask a question. The long-press gesture clearly differentiates it from sibling tools like tap_and_ask, swipe_and_ask, and screenshot_ask, even without naming them explicitly.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies the use case: perform a long-press and inspect the resulting screen via a screenshot and question. However, it does not explicitly state when to prefer this over tap_and_ask, swipe_and_ask, or other siblings, nor does it mention any exclusions.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Send an Android keyevent code (e.g. 4 = BACK, 66 = ENTER, 187 = APP_SWITCH). No vision call.
| Name | Required | Description | Default |
|---|---|---|---|
| keycode | Yes | Android KEYCODE_* integer value. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are present, so the description is the only source of behavior. It discloses the core action (sending a keycode) and that it does not use vision, but it does not clarify whether the key is pressed and released with a single event or describe timing/duration. Lacks details on side effects.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
One compact sentence with two clauses; the key action is front-loaded, and each part (action, examples, vision exclusion) adds value without redundancy.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Simple tool with one required parameter and no output schema. The description covers what it does and gives examples, so an agent can invoke it correctly. It does not explain return behavior or errors, but those are likely unnecessary for a fire-and-forget key event. Minor missing context about when to use it is covered under usage guidance.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The schema already documents keycode as an Android KEYCODE_* integer. The description adds specific example values (4=BACK, 66=ENTER, 187=APP_SWITCH), which clarify the range and meaning significantly beyond the schema's generic description.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States the specific verb 'send' and resource 'Android keyevent code', gives concrete examples distinguishing it from vision-based siblings like screenshot_ask and tap_and_ask, and explicitly notes 'No vision call.'
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Provides minimal guidance on when to use; the 'No vision call' implies it is not for visual tasks, but it does not explicitly name alternatives or conditions for selection. The examples imply use for system keys but lack explicit routing.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
For checking an ANIMATION or any effect that plays out over time (e.g. "does the strength indicator pulse smoothly?", "does the XP label fly up and fade out?", "does the sprite return to its start position?"). Optionally performs one action first (tap or swipe, or neither), then takes frameCount screenshots spaced intervalMs apart, and asks the SAME short question about each frame separately (each frame gets its own vision call, with its frame number in the prompt) - returns one answer per frame in order.
Sequential single-frame calls were chosen over sending several frames in one request: Runware's imageCaption does accept an undocumented multi-image array, and it works fine for exactly 2 frames, but degrades noticeably at 3+ (truncated/malformed answers in testing) - sequential calls are both more reliable and, per-frame, no more expensive. Keep frameCount modest (3-6) - each frame is a full separate vision call and cost scales linearly with it.
| Name | Required | Description | Default |
|---|---|---|---|
| x | No | Required for action=tap or action=swipe (swipe start x). | |
| y | No | Required for action=tap or action=swipe (swipe start y). | |
| x2 | No | Required for action=swipe (end x). | |
| y2 | No | Required for action=swipe (end y). | |
| action | Yes | Action to perform before starting the capture sequence. | |
| waitMs | No | Milliseconds to wait after the action before the FIRST screenshot (default 500) - same meaning as waitMs in tap_and_ask/swipe_and_ask, separate from intervalMs which spaces out the frames after that. | |
| question | Yes | The same short question asked about every captured frame (e.g. "Is the indicator visible? yes/no"). | |
| frameCount | Yes | How many screenshots to take, spaced intervalMs apart (2-8; keep modest, see description). | |
| intervalMs | Yes | Milliseconds between each screenshot (i.e. the sampling interval of the sequence). |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden, and it delivers: it reveals that each frame gets its own vision call with the frame number in the prompt, that answers come back one per frame in order, and that sequential calls were deliberately chosen over multi-image requests due to reliability degradation. This gives the agent accurate expectations about cost and behavior.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is longer than average, but every sentence contributes: purpose, action semantics, per-frame behavior, return ordering, rationale for sequential calls, and cost guidance. The key use case is front-loaded, and the engineering rationale is placed where it helps the agent decide rather than adding noise.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a tool with 9 parameters, no output schema, and no annotations, the description covers the core interaction fully: what triggers the sequence, what each frame does, how the question is applied, what the result order is, and cost implications. The remaining parameter details are already well documented in the input schema.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so the baseline is 3. The description adds extra value by advising frameCount stay modest (3-6), explaining linear cost scaling, and clarifying that waitMs is distinct from intervalMs and has the same meaning as in tap_and_ask/swipe_and_ask. This goes beyond the schema's structural descriptions.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description opens with a specific use case—checking an animation or effect that plays out over time—and clearly states the mechanism: perform an optional action, take frameCount screenshots spaced intervalMs apart, and ask the same question about each frame. This distinguishes it from single-shot siblings like screenshot_ask or tap_and_ask without needing to open their schemas.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
It explicitly says when to use the tool: for animations or time-based effects. It also explains when the optional action is tap, swipe, or neither. However, it does not explicitly name alternative tools or state when NOT to use this one, so the guidance is clear but not fully contrastive.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Take a screenshot of the current screen and ask a short question about it (e.g. "Is there an error dialog visible?", "How many word icons are on screen?", "What color is the strength indicator?"). Use this when you need to check state WITHOUT performing an action first. Phrase the question so a short answer is possible (yes/no, a number, a short label) - see this server's README "Cost model".
| Name | Required | Description | Default |
|---|---|---|---|
| question | Yes | A short, specific question about the current screen. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the burden of behavioral disclosure. It clearly communicates a non-action read-only behavior and implies the response is short ('so a short answer is possible'). It points to the README for cost details, adding context. While it doesn't describe the exact return format, the answer is implied by the question-asking purpose.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is concise (about 3 sentences) and front-loaded with the main action. Every sentence earns its place: statement of action, examples, usage guidance, and cost reference. No redundancy or filler.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a single-parameter, read-only tool without an output schema, the description is largely complete. It covers purpose, usage timing, question phrasing, and cost considerations. It could explicitly mention that the result is an answer to the question, but that is reasonably implied. The description is sufficient for an agent to call it correctly.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100% and the parameter description already clarifies the question. The tool description adds value by providing examples of valid questions and guidance on phrasing for short answers, which enriches the parameter's semantics beyond the schema alone.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description states a specific verb-resource pair ('Take a screenshot of the current screen and ask a short question about it') and provides clear examples. It also distinguishes itself from siblings by specifying 'WITHOUT performing an action first', which separates it from action-based tools like tap_and_ask or swipe_and_ask.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
It explicitly states when to use this tool ('when you need to check state WITHOUT performing an action first') and gives concrete guidance on phrasing questions for short answers. The reference to the README 'Cost model' provides additional usage context. This fully addresses when to use instead of alternatives.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Swipe (or drag, for drag-and-drop UIs) from (x1, y1) to (x2, y2), wait briefly, take a screenshot, and ask a short question about the result - all in one call.
| Name | Required | Description | Default |
|---|---|---|---|
| x1 | Yes | ||
| x2 | Yes | ||
| y1 | Yes | ||
| y2 | Yes | ||
| waitMs | No | Milliseconds to wait after the swipe before screenshotting (default 500). | |
| question | Yes | A short, specific question about the screen after the swipe. | |
| durationMs | No | Swipe duration in ms (default 300; use longer for drag-and-drop hold gestures). |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the transparency burden and does disclose the full ordered behavior: swipe, wait, screenshot, ask. It also notes the duration nuance for drag-and-drop holds. It doesn't detail coordinate units or what 'ask' returns, but the step sequence is clearly communicated.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
One sentence captures the entire workflow with no filler. The core action is front-loaded, and the drag-and-drop nuance is efficiently folded into the gesture description.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
The description is adequate for a simple composite gesture tool, but given no output schema and no annotations, it omits the return/answer semantics of 'ask', the coordinate system, and any cost or side-effect implications. These gaps prevent it from being fully self-sufficient.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is only 43%, so the description must compensate. It explains x1/y1/x2/y2 as the swipe's start and end points, and clarifies that question should be short and about the post-swipe screen. The optional waitMs and durationMs already have schema descriptions, and the prose adds the 'hold for drag-and-drop' nuance.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description states a specific verb+resource: swipe/drag from one coordinate to another, wait, screenshot, and ask a question. It clearly distinguishes itself from sibling tools like tap_and_ask and screenshot_ask by naming the gesture ('Swipe') and the compound workflow.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The instruction to use drag 'for drag-and-drop UIs' gives some contextual guidance, but there is no explicit when-to-use/when-not-to-use statement or reference to alternatives. The appropriate context is implied by the swipe gesture rather than directly contrasted with siblings.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Tap at device screen coordinates (x, y), wait briefly for the UI to react, take a screenshot, and ask a short question about the result - all in one call. Use this for any "tap here, then check what happened" step instead of calling separate tap/screenshot/ask tools.
| Name | Required | Description | Default |
|---|---|---|---|
| x | Yes | X coordinate in device pixels. | |
| y | Yes | Y coordinate in device pixels. | |
| waitMs | No | Milliseconds to wait after the tap before screenshotting (default 500). | |
| question | Yes | A short, specific question about the screen after the tap. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden of behavioral disclosure. It describes the sequence of actions (tap, wait, screenshot, ask) and mentions a wait period before screenshotting. However, it does not disclose what happens if the tap fails, the exact return format (e.g., does it return an image, a text answer, or both?), or any side effects like requiring a running app. The description is adequate for basic behavior but lacks depth on error handling or output, which is significant for a composite tool.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is two sentences with no redundancy. The first sentence front-loads the action sequence, and the second sentence provides direct usage guidance. Every word serves a purpose, and the structure is clean and immediately understandable. It avoids jargon and is well-scoped.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a composite tool with 4 parameters and no output schema, the description should cover the response format and any prerequisites. While it clearly states the purpose and usage, it does not mention what the tool returns (e.g., an answer to the question, a screenshot reference) or any necessary preconditions (e.g., the device being interactive). Since there is no output schema, the description's silence on return values leaves an agent without complete information for correctly interpreting the tool's outcome. This is a notable gap, so a 3.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the baseline is 3. The description adds minimal semantic value beyond the schema: it refers to 'wait briefly' which maps to waitMs, and characterizes the question as 'short and specific', but does not explain coordinate units or default wait behavior beyond what the schema provides. The description does not go beyond the schema definitions, so a 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's verb and resource: 'Tap at device screen coordinates (x, y), wait briefly for the UI to react, take a screenshot, and ask a short question about the result - all in one call.' It distinguishes itself from the siblings by defining its specific action (tap) and explicitly contrasting with calling separate tap/screenshot/ask tools. The title also reinforces the composite nature, so there is no ambiguity.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides explicit when-to-use guidance: 'Use this for any "tap here, then check what happened" step instead of calling separate tap/screenshot/ask tools.' This clearly defines the usage context and names an alternative (separate tools). It does not explicitly mention sibling tools like swipe_and_ask or long_press_and_ask, but the reference to 'tap' inherently implies a distinction from those. This is strong guidance, but not exhaustive about exclusions from all siblings, hence a 4.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Type text into whatever field currently has focus. No screenshot/vision call - pair with screenshot_ask if you need to confirm the result.
| Name | Required | Description | Default |
|---|---|---|---|
| text | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the behavioral burden. It discloses that typing targets the focused field and that screenshot/vision is not part of the operation, but it does not cover edge cases such as no focused field, whether existing text is replaced, or how special characters are handled.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two tight sentences: the core behavior is front-loaded, and the follow-up guidance about screenshot_ask earns its place. No filler or redundancy.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a one-parameter tool, the description is nearly sufficient: it states the target, the action, and the verification route. Missing failure-mode detail, such as what happens when no field has focus, keeps it from a 5.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0%, and the description does not add meaning beyond the property name 'text'. The parameter is simple, but nothing explains format, newline behavior, limits, or encoding, so the description fails to compensate for the schema's lack of documentation.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States a concrete action ('Type text') and a specific target ('whatever field currently has focus'), and explicitly warns against treating it as a screenshot/vision operation. This clearly differentiates it from screenshot_ask and the other _and_ask siblings.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Gives clear operational guidance: use it when a field has focus, and pair it with screenshot_ask when confirmation is needed. It does not explicitly contrast with press_key or other input tools, but the focus-based behavior is enough to guide selection.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Report the cumulative vision-provider spend for today and the configured alert/cap thresholds, without making any device or vision call.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full burden. It openly states 'without making any device or vision call', which discloses side-effect-free behavior, but it does not describe the output format, potential delays, or any other behavioral aspects. This is a reasonable disclosure but not comprehensive.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single, well-structured sentence that front-loads the verb and resource. Every word adds value, with no redundancy or filler.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a zero-parameter, no-output-schema tool, the description covers the essential information: what is reported and the guarantee of no side effects. It does not specify the return format or any prerequisites, but given the simplicity, it is sufficiently complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
There are zero parameters, so the baseline is 4. The description adds meaning about what the report contains (spend and thresholds) beyond the empty schema, which is exactly what is needed for a parameterless tool.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description uses the specific verb 'Report' and identifies the exact resource ('cumulative vision-provider spend for today') plus the alert/cap thresholds. It is clearly distinct from the sibling action-oriented tools (screenshot, tap, etc.) by stating it makes no device or vision call.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies the tool is for checking spend information and explicitly notes it does not make any device or vision call, but it does not name alternative tools or provide explicit when-to-use guidance. The use case is somewhat obvious given the sibling list, but the guidance is not explicit enough for a higher score.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v0.1.0logcat_greplong_press_and_askpress_keyrecord_and_askscreenshot_askswipe_and_asktap_and_asktype_textvision_spend_reportScored across 9 tools
Each tool has a clearly distinct purpose: screenshot_ask is passive state-checking, while tap/swipe/long_press_and_ask each combine a specific gesture with screenshot-and-ask. record_and_ask targets animations, type_text and press_key are direct input without vision, logcat_grep handles logs, and vision_spend_report tracks cost. No two tools overlap in function.
All tool names follow a consistent snake_case pattern, with a clear <action>_and_ask convention for vision-verifying interactions and simple verb_noun for the rest. The naming logically separates gesture tools from non-vision tools, making the set easy to navigate.
Nine tools is well-scoped for a mobile automation/verification server. Each tool addresses a concrete need—actions, verification, logging, cost monitoring—and none feel redundant or purely decorative. The count fits the domain without bloat or sparsity.
The tool surface covers the full cycle of mobile UI interaction and verification: direct input (type_text, press_key), gestures (tap/swipe/long_press), visual state checking (screenshot_ask, record_and_ask), log inspection (logcat_grep), and cost governance (vision_spend_report). No obvious dead ends or missing operations for the stated purpose.
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.
Cloud Android phones for AI agents: create a phone, read the screen, tap, type, install APKs, park.