Skip to main content
Glama

Clipform

by io.github.Clipform

Generate Video

clipform_generate_video
Destructive

Generate a video from images, video clips, or both, synced to an audio track. Use this for narrated question backgrounds, topic visualisations, or any form node that benefits from video. Combine with clipform_generate_tts for narrated audio and clipform_search_media for royalty-free images. Creates 9:16 (720x1280) with Ken Burns pan/zoom effects and transitions. Returns a public URL when complete.

Items: type "image" (Ken Burns motion) or "video" (cover-cropped, muted by default). Duration matches audio_url or set duration_seconds explicitly.

For multi-question builds, pass wait: false on every render: each call returns a job ID immediately, so all renders run in parallel - then collect URLs with clipform_check_render. Sequential waiting renders take 15-120 seconds EACH.

Choosing a render tool: for a recognisable form/quiz beat (guess-the-city, this-or-that, mystery reveal, multiple choice, photo montage...) reach for a video template first (clipform_list_video_templates + clipform_render_video_template) - it is a one-call recipe. Use clipform_generate_video for a narrated or audio-synced media montage (images/clips timed to a voice track). Use clipform_render_composition only when neither fits and you need a custom layer stack. Montage disambiguation: choose clipform_generate_video when the montage is narrated or synced to an audio track; choose the slideshow video template when it is silent (motion + transitions only, no voice-over). A render for a form node is not done until it is attached to that node. Pass node_id (and form_id) so the completed render attaches itself automatically - do not poll clipform_check_render to completion or manually chain clipform_upload_media_asset + clipform_attach_node_media; fire the render and move on.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
waitNotrue (default) blocks until the video is ready and returns its URL. false returns a job ID immediately - fire all renders first, then poll clipform_check_render. Use false whenever rendering more than one video.
itemsYesMedia items (images, video clips, or a mix)
contextYesDescribe the user's underlying goal in one sentence - not the tool you're calling.
duotoneNoTwo-tone editorial recolour on image items - desaturates then maps to a shadow->highlight palette. Pair with a halftone texture for a screen-print poster look.
form_idNoThe form UUID (required when node_id is set).
node_idNoForm node to attach this render to automatically once it completes - skips the manual clipform_upload_media_asset + clipform_attach_node_media steps and republishes the form if it's currently live. Requires form_id.
textureNoPrint-style pattern overlay on image items - makes stock imagery read as designed (screen-print dither look)
captionsNoWord-level captions from clipform_generate_tts - carried onto the attached media asset. Only used when node_id is set.
audio_urlNoAudio track URL. Video duration matches audio duration.
transitionNo
style_presetNoKen Burns style preset: cinematic, dramatic, calm, documentary, dreamy, moody, energetic
random_effectsNoShuffle Ken Burns effects across image items (default: true)
background_colorNoBackground color (default '#000')
duration_secondsNoVideo duration in seconds (required if no audio_url)
background_audio_urlNoAmbience/music bed under the narration (crowd noise, room tone). Loops to fill the video. Find tracks with clipform_search_music.
background_audio_volumeNoBackground bed volume 0-1 (default 0.15 - sits under speech)

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault
job_idNoPresent when status is 'rendering' - pass to check_render
statusYes'rendering' when wait:false (poll check_render); 'complete' with a public_url when wait:true
attachedNoTrue when node_id was provided and the render was attached to the node automatically (present once the attach outcome is known).
public_urlNoPresent when status is 'complete' - attach via upload_media_asset then attach_node_media (fit_media: true), unless node_id was set (auto-attached)
republishedNoTrue when the form was live and was republished to include this media.
attach_errorNoPresent when node_id was provided but auto-attach failed - the render itself still succeeded.
media_asset_idNoThe workspace media asset created from this render, when attached.
duration_secondsNo

TDQS

A4.6/5.0
Behavior4/5

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

Annotations declare readOnlyHint=false, openWorldHint=true, idempotentHint=false, destructiveHint=true. The description adds behavioral specifics: it creates a public URL, returns a job ID immediately when wait:false, attaches itself to a node automatically, and mentions that video clips are 'muted by default'. It does not contradict the annotations. The only slight gap is that it doesn't explicitly acknowledge the destructive nature (which the annotation covers), so it doesn't fully own the burden, but it adds valuable workflow context.

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?

The description is long but each section earns its place. It leads with the core purpose and key output characteristics, then provides guidance on items, rendering, tool selection, and attachment. The structure is logical: purpose → key behaviors → usage → disambiguation → integration. While slightly verbose, it packs in decision-critical information that prevents misuse, making the length justified.

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

Completeness5/5

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

Given the complexity (16 parameters including nested objects like items, duotone, texture, captions) and the existence of an output schema (indicated by 'Has output schema: true'), the description covers all essential aspects: what the tool does, how to configure it, how to handle asynchronous jobs, how to integrate with node attachments, and how to choose it over siblings. An agent would have no trouble invoking it correctly, even without prior knowledge of the platform.

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 description coverage is 94%, so most parameters already have descriptions in the schema. The description goes beyond that by explaining the relationship between items and duration ('Duration matches audio_url or set duration_seconds explicitly'), and by providing guidance on the wait parameter's impact on parallel rendering. It also clarifies the role of node_id for automatic attachment. While it doesn't detail every parameter, it adds meaningful workflow-level semantics that the schema alone doesn't convey, which justifies a score above the baseline of 3.

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?

The description opens with a clear, specific statement: 'Generate a video from images, video clips, or both, synced to an audio track.' It immediately names the resource (video), the action (generate), and the input types. It also distinguishes itself from sibling tools by explicitly naming clipform_render_video_template and clipform_render_composition as alternatives for different use cases, so an agent can easily tell them apart.

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?

The description provides exceptionally detailed usage guidance. It states when to use this tool ('narrated question backgrounds, topic visualisations'), gives explicit rules for parallel rendering ('pass wait: false on every render'), and includes a dedicated 'Choosing a render tool' section that contrasts with clipform_list_video_templates + clipform_render_video_template, and clipform_render_composition. It also clarifies the montage vs. slideshow distinction. This is far beyond a simple hint.

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

All Clipform toolsInspect
Try in Browser
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

TDQS

A3.9/5.0
Disambiguation3/5

Most form/node CRUD tools are clearly distinct, but the render/generate trio (generate_video, render_video_template, render_composition) and get_responses vs get_results can be confused by name alone. The descriptions include strong disambiguation guidance, so the overlap is manageable but still present.

Naming Consistency4/5

The vast majority of tools follow a consistent clipform_verb_noun snake_case pattern, with create/get/update/delete/list used predictably. Minor deviations include get_more_tools lacking the clipform_ prefix and whoami not matching the verb_noun convention.

Tool Count2/5

At 34 tools, this is above the range where a toolset feels well-scoped, and several entries are auxiliary or internal (whoami, log_generation, get_more_tools, search_news). The domain is broad, but the surface could be consolidated without losing core capabilities.

Completeness4/5

The set covers form and node lifecycle, media upload/attachment, logic wiring, publishing, responses/results, and content generation (TTS, video, music, stock media). Minor gaps remain, such as no conditional branching in set_logic and no general media-library listing tool.

Resources