Skip to main content
Glama

Create video

create_video

Make an AI video clip with sound from a prompt: any subject, action, setting, camera move (dolly, orbit, crane, handheld, drone), lens and mood, in a cinematic, documentary, product, social, anime or any look you describe. Start from a picture (it becomes the first frame) and optionally end on another for a controlled transition. Feature a saved character by writing its @tag. Choose the standard or premium engine (premium: longer single takes, the finest detail and more shapes, such as 1:1, 21:9 and 3:4), the shape (9:16 for Reels, TikTok and Shorts; 16:9), the length and the resolution up to 4k; get_capabilities lists what each engine offers this account. Clips render in one to a few minutes: poll get_job, or the PrismPoster card follows it. Keep working on a result with change_video, extend_video or upscale_video. Spends the account's PrismPoster credits at the same price as on prismposter.com (quote_job prices it first, free). When the person must decide something first (a price above their ask-first amount, consent for real people in pictures, the right to use a clip), it returns approval_required and nothing runs until they answer in PrismPoster's card or at the link. Everything made is saved to their PrismPoster library. Requests involving anyone under 18 or depicting public figures or office holders are refused before any charge.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "create_video"
  }
}

Input Schema

NameRequiredDescriptionDefault
engineNostandard (default) or premium: premium renders longer single takes at 1080p or 4k and costs more. get_capabilities says what this account can use.standard
promptNoWhat happens in the clip: subject, action, setting, camera, mood. Required unless image_url is given. Write it for the clip length; a brief longer than the clip holds is refused before any charge.
image_urlNoOptional https URL of a picture to start the clip from (it becomes the first frame).
resolutionNoOutput resolution, from those get_capabilities lists for the engine. Defaults to the engine's default.
approval_idNoOnly after an approval_required answer: the approval_id it returned, once the person has answered. The job then runs exactly as they approved it and the other arguments are ignored.
aspect_ratioNoFrame shape. 9:16 for Reels, TikTok and Shorts; 16:9 for YouTube; 1:1 for feeds; 21:9; 3:4. The standard engine may offer only some of these (get_capabilities lists the shapes of each engine). Defaults to 16:9.
end_image_urlNoOptional https URL of a picture to end the clip on. Requires image_url.
generate_audioNoPremium only: whether the clip gets generated sound. Defaults to on.
idempotency_keyNoOptional. A unique key for this job (8-128 letters, digits, . _ : -). Repeating a call with the same key returns the first call's job instead of starting and charging a second one, so set it whenever a retry is possible.
reference_audioNoOptional tracks whose rhythm and tone to follow: a track_id from get_job or an audio asset_id (list_library or import_media). For a clip from a prompt, not with image_url. Needs a plan or credit pack.
reference_clipsNoOptional clips whose motion, camera or feel to follow: a clip_id from get_job or a video asset_id (list_library or import_media). For a clip from a prompt, not with image_url. Needs a plan or credit pack.
duration_secondsNoClip length in seconds, from the lengths get_capabilities lists for the engine. Defaults to the engine's default length.
reference_imagesNoOptional pictures to match (a person, product, place or look), as https links: one made here or one import_media brought in. For a clip from a prompt, not with image_url.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • changedInput schema / properties / aspect_ratio / description
      Previous value: -"Frame shape. 9:16 for Reels, TikTok and Shorts; 16:9 for YouTube; 1:1 for feeds. Defaults to 16:9."New value: +"Frame shape. 9:16 for Reels, TikTok and Shorts; 16:9 for YouTube; 1:1 for feeds; 21:9; 3:4. The standard engine may offer only some of these (get_capabilities lists the shapes of each engine). Defaults to 16:9."
  2. First observed

TDQS

A4.5/5.0
Behavior5/5

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

The description adds substantial behavior beyond the annotations: it spends PrismPoster credits, can return approval_required with nothing running until the person answers, refuses under-18/public-figure requests before charging, renders in minutes, and saves results to the library. These are exactly the operational and safety details an agent needs for a credit-consuming mutation.

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 front-loaded with the core action and every major sentence contributes useful operational context. It is somewhat long and dense as a single paragraph, with elaborate prompting language, but not repetitive enough to waste space.

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?

For a complex 13-parameter, credit-spending creation tool with no output schema, it covers cost, approval workflow, refusals, async polling, library saving, and follow-up tools well. It omits the actual return shape of create_video and several optional reference parameters, but the schema fills those gaps.

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 100%, so the schema already explains all 13 parameters and the baseline is 3. The description still adds practical meaning for important choices: premium versus standard engine capabilities, aspect-ratio use cases, image_url as the first frame, end_image_url for controlled transitions, and @tag character reference; it does not discuss reference_audio, reference_clips, reference_images, approval_id, or idempotency_key, so it is not exhaustive.

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 states a specific verb and resource: making an AI video clip from a prompt, with sound and optional image start/end. It explicitly distinguishes continuing work via change_video, extend_video, or upscale_video, so an agent can tell this is the creation step rather than a follow-up edit.

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

Usage Guidelines4/5

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

It gives clear usage context: start from an image, end on another image, use an @tag for a saved character, choose an engine, poll get_job, and use sibling tools for follow-up work. It also names get_capabilities and quote_job as prerequisite checks, but it does not explicitly state when not to use this versus other creation tools such as create_image.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources