Skip to main content
Glama

Generate a video (beta)

generate_video

Beta. The AI director makes a motion video from a brief and/or up to 3 demos (their screens, copy and narration): it picks a look, writes a storyboard of 10–15 beats (hook, product moments, CTA), gathers screens, b-roll, music and sound effects, designs every scene and records the narration. It runs as a job (usually 1–3 minutes): this waits up to 20 seconds, then answers with the finished video or its progress — if it is still "generating", call get_video later instead of generating again. Uses AI credits (5 up to a minute, 10 longer; refunded if it fails) and AI voice characters for narration. A walkthrough keeps a demo step's recorded narration when its script stays word for word; other formats write their own script. On the Free plan the user's email must be verified first, and the brand kit styles one video: the first made from the workspace's first demo (list_brand_kits brandVideo); others get the built-in look.

Requires: demos:read, demos:write.

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": "generate_video"
  }
}

Input Schema

NameRequiredDescriptionDefault
lookNoArt direction; leave it out to let the director pick (list_video_blocks lists the looks).
mediaNoStock footage and AI images where they help.
musicNoA generated music bed.
proofNoProof the video may show word for word: numbers, customer quotes (with who said them), customer names. The director never invents these, so without proof it shows none.
titleNo
aspectNo16:9
formatNoteaser (punchy, 15–30 s), walkthrough (step by step on the real screens), launch (reveal + features), explainer (problem → how it works), social (vertical-friendly, huge type).teaser
promptNoWhat the video is about: product, audience, key message, tone, a call to action — or a full script.
demoIdsNoDemos to draw screens, copy and narration from (list_demos).
narrateNoAn AI voice reads the narration (Pro and Team).
lengthSecondsNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observed

TDQS

A4.9/5.0
Behavior5/5

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

Annotations only declare non-read-only, non-idempotent, non-destructive; the description goes far beyond by disclosing async job semantics, the 20-second wait, AI credit cost (5 credits under a minute, 10 longer) with refund on failure, and voice character consumption. It also flags the non-idempotent cost risk implicitly by warning against regenerating while a job is in flight.

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 beta flag and core action, then layers constraints; nearly every sentence carries operational information. It is a dense run-on paragraph, though, and the brand-kit/Free-plan caveats could be split into their own sentences for faster scanning.

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?

With 11 parameters, no output schema, and non-trivial async behavior, the description covers what the tool returns (finished video or progress status), costs, permissions, plan gating, and format nuances. An agent has everything needed to call it correctly and to avoid duplicate generation.

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

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 73% but the description still adds meaning: it explains that omitting `look` lets the director choose, ties `lengthSeconds` to credit tiers, explains which inputs feed screens/copy/narration (demos), and clarifies the walkthrough-versus-other-format narration behavior that no schema field documents.

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 specific verb and resource ('makes a motion video from a brief and/or up to 3 demos') and enumerates the pipeline it runs (storyboard, screens, b-roll, music, scene design, narration). It explicitly distinguishes itself from the sibling get_video by saying to call that one later rather than generating again.

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?

Gives concrete routing rules: this is a job that waits up to 20 seconds, and if status is still 'generating' the agent should call get_video instead of re-invoking generate_video. It also names prerequisites (Free plan email verification, brand kit scoping, required scopes) and the format-specific narration rule.

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.