Skip to main content
Glama

FXI Studio

by io.github.fueledximagination
OverviewDiscussionsAdmin

Generate a video clip

generate_video

Generates one video clip from a prompt and saves it to your FXI library. Optional start and end frames, reference images, a reference or source video (for example your own clip to edit with Gemini Omni) and reference audio. Bring your own files with create_upload_url or upload_from_url first. Spends credits. Returns a generation id; check with get_generation (usually 1-3 min; prefer lifecycle_status).

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
seedNoOptional seed for repeatable results.
modelNoVideo model from list_models. Defaults to "p-video-2" (fast, lowest cost); for dialogue, multi-shot or complex camera work the result suggests "seedance-2-mini".
outfitNoOptional outfit for this clip only (overrides the character's default Look). Modest, everyday clothing words.
promptYesWhat happens in the clip, including camera movement.
durationNoClip length in seconds. Up to 20, or up to the model's max_duration_seconds from list_models (Seedance 2.5: 30). Within that, each model adjusts to its own range.
model_idNoDeprecated alias of "model". Prefer "model".
resolutionNoFor example "720p" or "1080p". Models that list resolutions in list_models accept only those.
video_roleNoHow reference_video is used by Gemini Omni: "source" edits that clip, "previous" continues from it, "reference" (default) guides motion and style.
aspect_ratioNoFor example "16:9" or "9:16".
character_idNoOptional id of one of your saved characters (find ids with list_characters). Their locked Identity and default Look go into the clip word for word, with their reference images where the model accepts them.
estimate_onlyNoIf true, returns the estimated credit cost for this exact request (model, duration, resolution) and does NOT generate anything or spend any credits.
voice_consentNoRequired with reference_audio: true only after the user confirms they own the audio or have the rights to use it, including any voice, music or lyrics in it.
generate_audioNoAsk for sound on models that support it.
idempotency_keyNoOptional. Reuse the same key to retry safely: the earlier result is returned and nothing is charged twice.
reference_audioNoOptional reference audio (models that support it). Requires voice_consent. Public https URL, or an asset_url from create_upload_url, upload_from_url or upload_media_data (used as-is).
reference_imageNoOptional reference image. Public https URL, or an asset_url from create_upload_url, upload_from_url or upload_media_data (used as-is).
reference_videoNoOptional reference or source video, mp4 or webm (models that support it; Gemini Omni uses it as the clip to edit). Public https URL, or an asset_url from create_upload_url, upload_from_url or upload_media_data (used as-is).
last_frame_imageNoOptional end frame image. Public https URL, or an asset_url from create_upload_url, upload_from_url or upload_media_data (used as-is).
reference_imagesNoOptional up to 4 reference images (models that accept several). Public https URL, or an asset_url from create_upload_url, upload_from_url or upload_media_data (used as-is).
first_frame_imageNoOptional start frame image. Public https URL, or an asset_url from create_upload_url, upload_from_url or upload_media_data (used as-is).

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • changedInput schema / properties / model / description
      Previous value: -"Video model from list_models. Defaults to \"seedance-2-fast\"."New value: +"Video model from list_models. Defaults to \"p-video-2\" (fast, lowest cost); for dialogue, multi-shot or complex camera work the result suggests \"seedance-2-mini\"."
  2. First observed

TDQS

A4.3/5.0
Behavior4/5

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

Annotations already cover readOnly=false, openWorld=true, idempotent=false, destructive=false. The description adds important behavioral context: it spends credits, returns a generation id, typically takes 1-3 minutes, and recommends checking lifecycle_status. It also notes the need to upload files first. It does not mention idempotency_key or voice_consent requirements, which the schema covers.

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

Conciseness5/5

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

Front-loads purpose in the first sentence, then efficiently covers optional inputs, prerequisites, cost, and next steps. Every sentence is informative with no redundancy.

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?

For a complex 20-parameter tool with no output schema, the description provides the essential context: what it does, required uploads, cost implications, return value (generation id), and how to check status. The rich schema fills in parameter details, so the description is complete enough for an agent to call it correctly.

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

Parameters3/5

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

Schema description coverage is 100%, so the schema fully documents all parameters. The description only lists optional input categories (start/end frames, reference images, reference video, reference audio) without adding syntax, format, or usage details beyond what the schema provides. Baseline 3 is appropriate when schema does the heavy lifting.

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 ('Generates') and resource ('one video clip from a prompt'), and clarifies the output destination ('saves it to your FXI library'). Distinguishes from sibling generate_image by specifying video, and from compose_scene/frame_chain by focusing on a single clip from a prompt.

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?

Provides clear context for use: prerequisite upload via create_upload_url or upload_from_url, and follow-up check with get_generation. It also names optional input types. However, it does not explicitly compare to alternatives like compose_scene or frame_chain, nor state when not to use it.

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

All FXI Studio toolsInspect
Try in Browser
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources