Showoff.dev – AI Mockup Generator

Create a mockup project

create_project

Creates a new still-image or film mockup project in one call: devices arranged by a composition, reference shot or template, screenshots from URLs placed on their screens, a background palette derived from the screenshots, optional texts and extra edit_scene ops. Use it to start any image, App Store set or animated film; use create_launch_video for a music-synced launch video, duplicate_project for a variant of an existing project, and edit_scene to change a project that already exists. The result is reviewed before it is saved: if it has errors, nothing is created and you get the findings with fixes. Screenshots that are files on disk: create the project first, then run the returned curl commands. Returns project_id, editor_url, the media, the scene, the review and a preview sketch.

Input Schema

TableJSON Schema

Name	Required	Description	Default
`ops`	No	Extra edit_scene ops applied at the end, in the same format as the `ops` of edit_scene.
`kind`	No	Still image or animated film. Fixed after creation. export_image renders a film as a still (its preview moment); the video of a film is exported by the user in the editor — there is no video export here. For a video made with you, use create_launch_video.	image
`name`	Yes	What the user finds the project by in their overview: "<app or product> · <what it shows>", in the user's language, e.g. "Tempo · App Store set, 5 images" or "Wetter · Das Wetter auf einen Blick". Never "Mockup", "Untitled" or "Test". Rename it later with edit_scene {op:"rename"} when the content changes.
`texts`	No	Up to three texts, laid out automatically, e.g. a headline and a subline. In a store set add titles with the add_text op and its `image` instead.
`folder`	No	Only when the user names a folder: the folder in their Showoff projects to put this in, e.g. "Launch videos" or a subfolder "Showoff/Trailer" (two levels at most). Folders are found by name and created when missing. "" moves it back to the top level.
`format`	No	Preset (quer, klassisch, quadrat, feed, hoch, appstore, banner, vierdrei, hochkant) or {width,height}.	quer
`devices`	No	The devices on stage, in order, at most 6. Leave it out when starting from a template to keep its devices.
`palette`	No	Palette id; default: the best-rated suggestion for the screenshots.
`reference`	No	Measured reference shot id (list_compositions) — copies its camera, format and device placement exactly. Overrides composition and format.
`composition`	No	Composition id; default is chosen from the devices.
`screenshots`	No	Screenshots by public URL (or tiny base64), each shown on the device at `device_index`. Files on disk: leave this out and run the curl commands the result returns.
`template_id`	No	Start from a template (list_templates) instead of an empty stage. Its devices, camera, background and motion come along; without `devices`, `device_index` of a screenshot counts the template's devices in order.

Schema Changelog

Changes observed during successful MCP inspections.

First observedOct 1, 2026

TDQS

A4.5/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations only declare readOnly=false, destructive=false and openWorld=true, so the description carries most of the behavioral burden and does well: it discloses the pre-save review ('if it has errors, nothing is created and you get the findings with fixes'), the disk-file/curl workflow, and the return payload. It stops short of stating auth/permission requirements or rate/limit behavior, leaving a small gap.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loads the one-call creation scope, then routing, then the review behavior, then the disk-file caveat and the return shape. Dense but every clause is functional; the long compound first sentence is the only minor cost to scanability.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Covers the creation scope, sibling routing, the destructive-failure review behavior, the disk-file path and the return fields, which is strong for a 12-parameter, mutation, no-output-schema tool. Missing only permission/auth notes and pagination-like limits, which are minor for this operation.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

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 nonetheless adds cross-parameter semantics beyond the schema: how reference overrides composition/format, that device default is iPhone unless asked otherwise, and the template_id interaction with devices and screenshot device_index.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a precise verb+resource ('Creates a new still-image or film mockup project in one call') and enumerates the constituent parts it composes (devices, composition, screenshots, palette, texts, ops). It explicitly distinguishes itself from siblings by naming create_launch_video, duplicate_project and edit_scene with the conditions that select them.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicit routing: 'use create_launch_video for a music-synced launch video, duplicate_project for a variant of an existing project, and edit_scene to change a project that already exists.' It also gives a within-tool branch (files on disk → create first, then run curl commands), covering both when-to-use and when-not.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All Showoff.dev – AI Mockup Generator tools Inspect

Try in Browser

Shareable, User-Owned Context BackendPeople, applications, and agents read, write, and coordinate together in one context lake. Reach it over MCP, a CLI, a Python SDK, or REST.fulcradynamics.comAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?