Ein MCP-Server für KI-Codierungsagenten (Claude Code, Cursor usw.) zum Testen von Android-Apps, deren gesamte Benutzeroberfläche in einer einzigen opaken OpenGL/Vulkan/Metal-Oberfläche gezeichnet wird – Cocos2d-x, Unity, Unreal, rohes OpenGL, libGDX und ähnliche Engines.
adb shell uiautomator dump und jedes auf dem Accessibility-Baum basierende Automatisierungstool (einschließlich der meisten MCP-Mobilautomatisierungsserver) funktionieren, indem sie die native Android-View-Hierarchie untersuchen – Schaltflächen, Beschriftungen, deren Text und Koordinaten. Das funktioniert hervorragend für eine normale Android-Benutzeroberfläche, die aus nativen Views aufgebaut ist.
Es funktioniert nicht für ein Spiel oder eine App, die ihre gesamte Benutzeroberfläche als Texturen in einer einzigen GLSurfaceView rendert. Aus Sicht des Accessibility-Baums gibt es genau eine opake View auf dem Bildschirm ohne Kinder, ohne Beschriftungen, ohne Koordinaten für irgendetwas darin. Es gibt nichts zu untersuchen – der Bildschirm ist eine Blackbox, egal wie viel UI tatsächlich darauf ist.
Der einzige verbleibende echte Beobachtungskanal sind Screenshots. Dieser Server ist um diese Tatsache herum aufgebaut, als Normalfall, nicht als gelegentlicher Fallback.
mobile-next/mobile-mcp ist der Allzweck-MCP-Mobilautomatisierungsserver und eine gute Standardwahl für normale native Apps: zuerst Accessibility-Baum (schnell, günstig, kein Vision-Modell, keine Bild-Tokens), nur wenn der Baum nicht liefert, was benötigt wird, wird auf Screenshots + Koordinaten zurückgegriffen.
Für eine OpenGL-Canvas-App ist dieser Fallback nicht gelegentlich – es ist der einzige Weg, der jemals funktioniert, jedes Mal. mobile-mcp-opengl ist speziell für diesen Fall gebaut und trifft dadurch zwei unterschiedliche Designentscheidungen:
Überhaupt kein Versuch des Accessibility-Baums. Es gibt nichts zu gewinnen, wenn man es versucht – er kommt für diese Apps immer leer zurück –, daher geht jedes Tool hier direkt zu Screenshot + Vision.
Die Vision-Analyse läuft über einen pluggable, separaten Provider (siehe unten), nicht über das Modell, das den aufrufenden Agenten ausführt. Eine funktionale QA-Schleife über ein Spiel kann leicht Hunderte von Screenshot-Prüfungen pro Sitzung erreichen; das alles über die eigene Vision des Haupt-Codierungsagenten zu leiten kostet echtes Geld und Tokens/Kontext, die du lieber für die eigentliche Codierungsarbeit ausgeben würdest. Hier gelangen die Screenshot-Bytes nie in den Kontext des aufrufenden Agenten – nur die kurze Textantwort des Providers.
Related MCP server: Android-MCP
Ein naives Design stellt tap, screenshot und ask als drei separate Werkzeuge bereit. Das zwingt den aufrufenden Agenten, für jede einzelne Interaktion eine mehrstufige Schleife zu orchestrieren: Tippen → Screenshot machen → an einen Vision-Schritt übergeben → Ergebnis lesen → entscheiden, was als Nächstes zu tun ist. Jeder dieser Schritte ist ein separater Tool-Aufruf und eine separate Runde – Tokens werden für Koordination verbrannt statt für die eigentliche Testlogik, und es gibt mehr Angriffsfläche für den Agenten, einen Schritt auszulassen, sie falsch zu ordnen oder über veralteten Zustand zwischen Aufrufen nachzudenken.
Stattdessen stellt dieser Server kombinierte Werkzeuge bereit – tap_and_ask, swipe_and_ask, long_press_and_ask – die die Aktion ausführen, kurz warten, den Screenshot machen, den Vision-Provider fragen und eine kurze Antwort zurückgeben, alles als ein einziger Tool-Aufruf. Ein mehrstufiges Testszenario kostet am Ende ungefähr eine Agentenrunde pro sinnvoller Prüfung, nicht drei oder vier.
Einfache screenshot_ask (nur beobachten, keine Aktion) und günstige Nicht-Vision-Werkzeuge (type_text, press_key, logcat_grep) sind ebenfalls für die Teile eines Testablaufs verfügbar, die dieses Muster nicht benötigen.
Tool | Was es tut | Vision-Aufruf? |
| Screenshot, dann eine kurze Frage dazu stellen | Ja |
| Tippen (x, y), warten, Screenshot, fragen | Ja |
| Wischen/Ziehen (x1,y1)→(x2,y2), warten, Screenshot, fragen | Ja |
| Langes Drücken (x, y) für eine Dauer, warten, Screenshot, fragen | Ja |
| Optionale Aktion, dann N Screenshots mit zeitlichem Abstand, dieselbe Frage zu jedem Frame stellen | Ja (N Aufrufe) |
| In das aktuell fokussierte Feld tippen | Nein |
| Ein Android- | Nein |
| Aktuelles logcat lesen, optional nach Regex gefiltert | Nein |
| Heutige kumulative Vision-Ausgaben und Schwellenwerte melden | Nein |
Bevorzuge logcat_grep gegenüber einem Vision-Aufruf, wenn das, was du brauchst, bereits in einer Log-Zeile steht (Abstürze, eigene Debug-Ausgaben, Netzwerkfehler) – es ist kostenlos und exakt, ein Vision-Aufruf ist beides nicht.
record_and_askEinzelbild-Werkzeuge können dir nicht sagen, ob etwas animiert korrekt funktioniert (pulsiert der Stärkeindikator sanft, fliegt ein Label hoch und blendet aus, springt ein Sprite zurück zu seiner Startposition). record_and_ask führt eine optionale Aktion aus (Tippen oder Wischen, oder keines), wartet waitMs (gleiche Bedeutung wie waitMs in tap_and_ask/swipe_and_ask – Zeit für die UI, um zu reagieren, bevor das erste Bild aufgenommen wird), erfasst dann frameCount Screenshots im Abstand von intervalMs und gibt eine kurze Antwort pro Frame zurück – der aufrufende Agent erhält eine Zeitleiste in einem Tool-Aufruf, anstatt selbst N separate Screenshot+Ask-Roundtrips zu orchestrieren.
Warum ein Vision-Aufruf pro Frame, nicht ein Aufruf mit allen Frames gebündelt. Runwares imageCaption akzeptiert offenbar ein undokumentiertes inputImages-Array (Plural) neben dem dokumentierten einzelnen inputImage – direkt gegen die API getestet. Es funktioniert sauber für genau 2 Bilder (ein Vorher/Nachher-Vergleich in derselben Anfrage kam korrekt und kohärent zurück). Bei 3+ Bildern in einer Anfrage erzeugten sowohl dieses Array-Parameter als auch ein manuell zusammengesetztes Side-by-Side-„Filmstreifen“-Bild in Tests abgeschnittene oder fehlerhafte Antworten – das kleine 7B-Vision-Modell verliert offenbar die Kohärenz, wenn die kombinierte visuelle+Anweisungslast in einem Aufruf zu hoch wird. Sequentielle Einzelbild-Aufrufe (der Ansatz dieses Tools) waren bei jeder getesteten Frame-Anzahl zuverlässig und sind nicht wesentlich teurer: Die Kosten werden von der Antwortlänge (siehe unten) dominiert, nicht von der Anzahl der Aufrufe, daher kosten N kurze sequentielle Antworten ungefähr so viel wie oder weniger als eine lange Multi-Bild-Antwort. Wenn dein eigener Provider Multi-Bild-Anfragen zuverlässiger verarbeitet, ist dies ein offensichtlicher Ort zum Optimieren – siehe „Bring dein eigenes Modell“.
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)Erfordert adb im PATH (oder ADB_PATH in .env gesetzt) und ein laufendes/verbundenes Gerät oder einen Emulator. Wenn mehr als eines angeschlossen ist, setze ADB_DEVICE_SERIAL (siehe adb devices).
Füge eine .mcp.json in deinem Projektstamm hinzu (diese Datei ist normalerweise projektspezifisch und git-ignoriert, da sie normalerweise auf einen maschinenspezifischen Pfad zeigt oder maschinenspezifische Env-Überschreibungen enthält):
{
"mcpServers": {
"mobile-opengl": {
"command": "node",
"args": ["/absolute/path/to/mobile-mcp-opengl/src/server.js"]
}
}
}Claude Code übernimmt dies automatisch für das Projekt. Der Server liest seine eigene .env (neben package.json in diesem Repo) für die gesamte Konfiguration – der aufrufende Agent muss selbst nie einen API-Schlüssel kennen oder übergeben.
Die Antwortlänge treibt die Kosten, nicht die Bildgröße. Dies wurde empirisch gegen den Standard-Provider Runware/Qwen2.5-VL-7B-Instruct gemessen: Dieselbe Frage mit einer erzwungenen Ein-Wort-Antwort kostete gleich viel ($0.0006) über Bildgrößen von 360×360 bis 1600×2400 (Retina-Klasse). Dasselbe 1024×1024-Bild mit einer offenen „Beschreibe dies“-Aufforderung kostete $0.0013–0.0019 – 2-3x mehr – rein weil das Modell eine längere Antwort schrieb, nicht weil das Bild größer war.
Praktische Auswirkungen:
Vergiss das Herunterskalieren von Screenshots, bevor du sie sendest – es reduziert die Kosten für diesen Provider nicht wesentlich und du verlierst Details, die du vielleicht brauchst.
Formuliere Fragen immer so, dass kurze Antworten erzwungen werden: ja/nein, eine Zahl, ein kurzes Label, ein winziges JSON-Objekt mit ein paar Feldern. Jedes Tool in diesem Server fügt automatisch eine Kurzantwort-Anweisung hinzu, aber eine vage offene Frage („Was siehst du?“) kann das Modell dennoch zu einer längeren Antwort drängen als eine spezifische („Ist der Fehlerdialog sichtbar? ja/nein“).
Bei ~$0.0006/Aufruf für gut formulierte kurze Fragen kostet eine 500-Aufruf-QA-Sitzung ungefähr $0.30. Das gleiche Volumen an offenen „Beschreibe den Bildschirm“-Fragen kann das 2-3x kosten.
Jeder Vision-Aufruf wird in .vision-log.jsonl protokolliert (JSONL, ein Eintrag pro Aufruf: Zeitstempel, Frage, Antwort, Kosten). Zwei unabhängige Schutzmechanismen sitzen auf diesem Log, beide provider-agnostisch (sie arbeiten mit dem, was ein Provider als costUsd meldet):
Pro-Aufruf-Warnung (VISION_ALERT_USD, Standard $0.0015): Wenn ein einzelner Aufruf darüber zurückkommt, enthält die Antwort des Tools einen [COST ALERT]-Hinweis, der dir sagt, dass das Modell wahrscheinlich die Kurzantwort-Anweisung ignoriert hat – ein Signal, die Frage umzuformulieren, nicht etwas, das man stillschweigend hinnehmen sollte.
Tageslimit (VISION_SESSION_CAP_USD, Standard $2.00): Sobald die heutigen kumulierten protokollierten Ausgaben diesen Wert erreichen, wird jeder weitere Vision-Aufruf komplett abgelehnt (bevor er den Provider erreicht), bis das Limit erhöht wird oder der Tag umschlägt. Dies ist ein harter Stopp gegen eine außer Kontrolle geratene Schleife, nicht nur eine Warnung.
Rufe vision_spend_report jederzeit auf, um die heutige Summe zu prüfen, ohne einen Geräte- oder Vision-Aufruf zu machen.
Wenn ein Provider keine Kosten melden kann (siehe openai-compatible unten), werden Aufrufe von ihm mit costUsd: null protokolliert und lösen nie die Warnung aus oder zählen zum Limit – die Schutzmechanismen können einfach keine Ausgaben schützen, die sie nicht sehen können.
Die Vision-Analyse läuft über src/providers/visionProvider.js, das einen Provider anhand des Namens aus VISION_PROVIDER in .env auswählt. Zwei sind integriert:
runware (Standard) – spricht direkt mit der imageCaption-Aufgabe von Runware.ai und verwendet standardmäßig Qwen2.5-VL-7B-Instruct (AIR-ID runware:152@2). Runware und OpenRouter sind zwei separate Dienste mit separaten API-Schlüsseln und Modellkatalogen – dies spricht direkt mit Runware, nicht über OpenRouter.
openai-compatible – ein generischer Provider für alles, was das OpenAI-Chat-Completions-Vision-Format spricht (image_url-Inhaltsteile). Funktioniert mit OpenRouter, einem lokalen Ollama/LM-Studio-Server mit einem Vision-Modell, Groq, Together.ai oder jedem anderen kompatiblen Endpunkt. Konfiguriere OPENAI_COMPATIBLE_BASE_URL, OPENAI_COMPATIBLE_API_KEY, OPENAI_COMPATIBLE_MODEL in .env. Die meisten OpenAI-kompatiblen APIs melden Token-Nutzung statt eines festen Dollar-Betrags; setze OPENAI_COMPATIBLE_PRICE_PER_1M_INPUT/_OUTPUT, wenn du möchtest, dass dieser Provider costUsd daraus schätzt (andernfalls sind Kostenverfolgung/Schutzmechanismen für diesen Provider wirkungslos, wie oben erwähnt).
Um einen vollständig benutzerdefinierten Provider hinzuzufügen (ein selbst gehostetes Modell, eine völlig andere API-Form), kopiere src/providers/openaiCompatibleProvider.js als Ausgangspunkt und implementiere:
async function ask(imageBuffer, mimeType, question) {
// return { text: string, costUsd: number | null }
}
module.exports = { ask };und registriere ihn mit einem Namen in loadProvider() von src/providers/visionProvider.js.
MIT
Entwickelt von 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.