Create and preview self-contained web apps in iPhone, Duo, iPad, and Headless views.
Scored across 2 tools
open_ipadian and preview_app have clearly distinct targets: one opens the iPadian home environment, the other renders a specific HTML app inside it. Their descriptions make the boundary obvious, with no overlapping purpose that would cause misselection.
Both tool names follow a consistent snake_case verb_noun pattern: open_ipadian and preview_app. There is no mixing of conventions or vague action words.
Two tools is borderline thin for a server meant to support the iPadian app environment. While each tool is meaningful, the set feels underpowered for managing or interacting with apps beyond opening and previewing.
The surface covers opening the environment and rendering an unsaved draft, but it lacks any read, list, save, update, or delete operations for the local app library. This leaves significant gaps for an agent trying to manage apps or verify saved state.
Open the iOS-style iPadian home screen with app icons, a dock, App Store, Create, and My Apps. After opening, briefly explain in chat that users can tap apps, browse App Store for ready-made apps or choose Create for examples, or ask your assistant to build and revise their own web app. Explain that Save locally keeps app code in this browser, not runtime progress. This tool cannot read the local library.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true, destructiveHint=false, and openWorldHint=false, and the description is consistent with them. It adds real behavioral context beyond the annotations: 'This tool cannot read the local library' is a concrete capability boundary, and it clarifies that 'Save locally' stores app code in the browser rather than runtime progress.
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?
Purpose is front-loaded in the first sentence and each following sentence adds distinct information (post-open guidance, storage semantics, capability limit). It is on the long side and drifts into chat-scripting territory, but nothing is redundant.
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, read-only UI-opening tool with no output schema, the description covers what opens, what the user can do next, and the key limitation. Nothing an agent needs to invoke it correctly appears to be missing.
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 tool takes zero parameters, so the baseline of 4 applies; there are no parameter semantics the description needs to compensate for.
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 verb+resource ('Open the iOS-style iPadian home screen') and enumerates its contents (app icons, dock, App Store, Create, My Apps), so the agent knows exactly what surfaces. It does not differentiate itself from the sibling preview_app, which is the only missing piece.
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 is mostly post-invocation scripting ('After opening, briefly explain...'), which implies the tool is the entry point, but it never states when to call it versus preview_app or any precondition. Usage is implied rather than spelled out.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Render a complete self-contained HTML app in iPadian. Include inline CSS and JavaScript; external requests, libraries, browser storage and navigation permissions are unavailable inside the sandbox. Keep the same id when revising an app. This returns an unsaved draft; only the user’s Save locally button stores it. Never claim the app was saved or published.
| Name | Required | Description | Default |
|---|---|---|---|
| id | Yes | ||
| code | Yes | ||
| name | Yes | ||
| duoPosture | No | unfolded | |
| previewMode | No | duo |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations only declare readOnlyHint/openWorldHint/destructiveHint; the description goes well beyond that by disclosing that external requests and browser storage are unavailable, that output is an unsaved draft requiring the user's Save locally button, that id reuse is expected on revision, and by warning the agent never to claim the app was saved. The 'unsaved draft' framing is consistent with the read-only annotation rather than contradicting it.
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?
Five short sentences, each carrying distinct information, with the core purpose front-loaded and the constraints following in priority order. The final 'never claim saved or published' line slightly restates the draft point but functions as an important behavioral guardrail, so it still 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 no output schema and thin annotations, the description adequately covers sandbox limits, draft semantics, id reuse, and expected inlining. It does not say what a failed render returns or how the preview modes change behavior, which are the remaining gaps for a five-parameter tool.
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% for five parameters, so the description carries the full burden and only partly delivers: it implies the id (reuse on revision) and constrains code to inline CSS/JS, but name, previewMode and duoPosture are never explained, nor is the enum meaning of the latter two.
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 concrete verb+resource: 'Render a complete self-contained HTML app in iPadian', which tells an agent exactly what the tool produces. It is clear and specific, but it never mentions the sibling open_ipadian, so differentiation must be inferred from the tool names alone.
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?
Usage context is strong: it states the sandbox limitations (no external requests, libraries, storage, navigation), instructs to keep the same id when revising an app, and clarifies the draft lifecycle. It stops short of explicit when-not-to-use routing against open_ipadian, so it sits just below the top 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.
open_ipadianpreview_appCreate, edit, preview, and deploy full-stack web apps to a live URL from any MCP client.
Let your AI agent build, preview and update web apps, download source and publish when you allow it.
Make your own apps and tools: build, edit and publish web apps from a conversation.
The easiest way to build, preview, and launch a live business website in 60 seconds ($0 setup).
Add one secure layer between your agents and this server.
