Skip to main content
Glama
QIANLING-0831

Flow Workspace MCP

by QIANLING-0831
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Local

Generate and Download a Google Flow Video

flow_generate_video
Destructive

Submit one Google Flow video generation job, return a persistent job ID to poll, then optionally upscale and download outputs to a local directory.

Instructions

THE REQUIRED AND EXCLUSIVE PATH for every Google Flow video request. Submits one persistent job, waits briefly, and returns. If status=processing, poll flow_job_status with the SAME job ID; never generate again. It optionally upscales and downloads locally. Never open or control Flow with browser/computer-use tools. Requires explicit credit authorization.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
modelNoNormalized video model ID returned by flow_inspect_account, e.g. omni-flash, veo-3-1-lite, veo-3-1-fast, or veo-3-1-quality. Exact visible labels are also accepted. Use ui-default to keep the selected model.ui-default
promptYesVideo prompt in any language, describing subject, action, setting, camera, lighting, style, and audio as desired.
outputsNoNumber of generated video outputs requested from Flow. Credits are typically charged per generation.
upscaleNoExact normalized upscale ID returned by flow_inspect_account (for example 1080p, 2x, or 4k), none, or highest_available. Unsupported and unavailable choices fail explicitly.none
downloadNoWhen true, download ready outputs immediately. When false, leave them in Flow and return the job.
fileNameNoOptional safe file stem. The job ID and downloaded extension are added automatically.
accountIdNoOmit accountId to use the most recently verified connected account. Supply it only when the user explicitly chose a different connected account returned by flow_list_accounts.
aspectRatioNoExact video aspect ratio returned by flow_inspect_account, or ui-default.ui-default
flowProjectNoExisting Flow project name to open. If omitted, the current project is reused or a new project is created.
referenceFilesNoAbsolute paths to optional local images or videos to attach as Flow references/ingredients/frames.
timeoutSecondsNoMaximum seconds to wait for this generation or upscale. A timeout leaves a persistent job that can be polled.
durationSecondsNoRequested clip length only when flow_inspect_account reports that exact value in visibleDurations. Omit when the current Flow Agent UI exposes no duration control.
outputDirectoryYesAbsolute directory where downloads and .flow.json manifests are saved, e.g. C:\project\public\generated\flow.
confirmCreditSpendYesMust be true. Confirms the user explicitly authorized this operation to consume Google Flow/AI credits.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

A4.4/5.0
Behavior4/5

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

Annotations already declare destructiveHint=true and openWorldHint=true, so the safety profile is covered. The description adds genuinely new behavior: one persistent job, a brief wait then return, timeout leaving a pollable job, and credit spend requiring authorization. It stops short of describing the returned job/manifest shape, which the missing output schema would otherwise require.

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?

Four tight sentences, front-loaded with the exclusivity claim and lifecycle, then routing, then the prohibition. Every sentence carries a distinct instruction with no filler.

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 14-parameter, credit-consuming, non-idempotent tool with no output schema, the description covers the job lifecycle, the polling escape hatch, the timeout failure mode, and the authorization gate. It could say a bit more about what is returned on the initial call (job ID/manifest), but the essentials are present.

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 each of the 14 parameters is already documented in the schema (model IDs, upscale IDs, durationSeconds visibility rules, etc.). The description adds no parameter-level detail beyond that, so the baseline 3 applies.

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 ('submits one persistent job' for a Google Flow video) and explicitly positions itself as 'THE REQUIRED AND EXCLUSIVE PATH for every Google Flow video request,' which cleanly separates it from flow_generate_image and flow_upscale_video. The submit-wait-return lifecycle is named up front.

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 when/when-not guidance: poll flow_job_status with the SAME job ID on status=processing, 'never generate again,' and 'Never open or control Flow with browser/computer-use tools.' It also names the precondition (explicit credit authorization), leaving almost nothing to inference.

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

Deploy Server

Other Tools