NewMCP · November 2026 Release

Rask MCP Server: Run AI Video Localization From an Agent

Rask MCP connects AI video dubbing, translation, voice cloning and lip-sync to the assistant you already use. Ask in your own words, approve the price, and the finished videos come back in the same chat.
Included on all paid plans
A chat with Rask: someone asks to dub three product videos into Spanish, German and Japanese in their own voice, Rask replies that it will use 36 minutes and shows the balance before and after with an Approve button, then returns nine finished versions
Trusted by global teams
How it works

AI video dubbing and translation, from one message

Translating one video is easy. Keeping a library of them in five markets is a production problem. The Rask MCP server hands that video localization work to the assistant your team already talks to.
How Rask MCP works: a prompt asking to dub a video into Spanish, German and Japanese and keep the speaker's voice and lip sync, passed through ChatGPT, Claude, Cursor or VS Code connected to Rask, and returned as three finished videos with their language flags
Use cases

How teams use Rask MCP

The same connection, four different working days. Every example below is a real Rask localization.
Play video
A creator video translated and dubbed by Rask AI, with the original voice kept

A weekly episode, in three more languages

Creators publish an episode in three more languages without leaving the chat where they write the script. The assistant sends the file, dubs it in their own voice and saves every version beside the original.
Play video
A marketing video translated by Rask AI with translated subtitles

Every market on the day of the launch

One request returns the dub, the lip-sync and the subtitles for each language, so a campaign video reaches every market on launch day. The team asks for it in the assistant it already uses for copy, so no new tool enters the process.
Play video
A training video dubbed by Rask AI into several languages

A whole course, one request

A full onboarding course is localized in a single request, with the company glossary applied so product and process terms read the same way in every module. When a policy changes, the assistant edits the affected lines and records only those again.
Play video
A media video localized by Rask AI into several languages

A client note becomes a fix, not a re-delivery

A note in plain language becomes a terminology fix across the delivery: the assistant finds the lines, corrects them and records only what changed, so the approved track stays as it was. The review happens in the chat, which keeps a round of changes short.
Spend control

The price is agreed before the work starts

An assistant that can spend your minutes needs a limit it cannot talk its way around. Rask puts one on every paid step.

It asks the price first

Before anything paid, Rask returns the exact cost in minutes and the balance you would have left. Asking costs nothing.

You approve it

The price is tied to that exact job. In Claude Code, Rask shows it in a dialog only a person can answer, so the assistant cannot approve on your behalf.

It never costs more

The approved amount is a ceiling. If the price moved, the job stops instead of charging more, and a repeated request never charges twice.
Review

Change one line, redub that line

The first version is a draft, and your team decides when it is finished. Say what is wrong the way you would to a colleague, and only those lines are recorded again. The rest of the track stays as approved.
The earlier version is kept, so a change is never a risk. Add your terms once and they stay the same across the whole library.
A chat with Rask: someone asks to replace the word cart with carrito in the Spanish version, Rask shows the seven lines it found and offers to re-record only those with the redub included and no minutes taken, then confirms it is done and the earlier version is saved
The 29 actions

Everything your AI agent is allowed to do

You decide what the connection may touch. An assistant you connect without spending access can plan and price a job, and never pay for one.

Your account

Check the plan, the balance and the languages available
rask_get_accountrask_get_usagerask_list_languages

Getting video in

Send a file from your computer, or import a link
rask_create_uploadrask_import_media

Projects

Start a localization, see what is running, collect the finished files
rask_create_projectrask_list_projectsrask_get_projectrask_get_download_links

The script

Read the transcript, change wording, translate parts again
rask_get_transcriptrask_edit_transcriptrask_retranslate_segments

Versions

Save a version before a round of edits and go back to it
rask_list_versionsrask_save_versionrask_restore_version

Voices

Hear the options, assign a voice to each speaker, create a clone
rask_get_voice_setuprask_list_voicesrask_set_speaker_voicerask_create_voice_clone

The work itself

Quote the price, dub, lip-sync, translate on-screen text, or do all of it at once
rask_estimate_costrask_start_dubrask_start_lipsyncrask_start_onscreenrask_start_localizationrask_generate_segments

Progress

Check how a long job is going, or stop it
rask_get_jobrask_cancel_job

Everything else

Look up and read anything else in the Rask API
rask_api_operationsrask_api_read
Three ready-made walkthroughs come with it: localize a video, edit and redub, and add lip-sync.
For developers

The video localization API behind the MCP server

Rask MCP runs on the public Rask API v3. Ask for a job in the chat, or make the same call from your own code when it has to run every week without you.
# pip install rask-sdk
from rask_sdk import Rask

rask = Rask()                                   # reads RASK_API_KEY

# the price first - free, and it changes nothing
estimate = rask.projects.estimate_project(
    media_id=media_id, target_language="es-es"
)

# the price you accepted is the ceiling: a higher one fails instead of charging
created = rask.projects.create_project(
    media_id=media_id,
    target_language="es-es",
    expected_cost_minutes=estimate.total_minutes,
)
processed = rask.jobs.wait(created.id)

outputs = rask.exports.list_project_outputs(processed.result.project_id).data
print(next(o.url for o in outputs if o.kind == "dubbed_video"))
// npm install @braskai/api
import { Rask } from '@braskai/api';

const rask = new Rask();                        // reads RASK_API_KEY
const order = { media_id: mediaId, target_language: 'es-es' };

// the price first - free, and it changes nothing
const { data: estimate } = await rask.projects.estimateProject({ body: order });

// the price you accepted is the ceiling
const { data: created } = await rask.projects.createProject({
  body: { ...order, expected_cost_minutes: estimate.total_minutes },
});
const { data: processed } = await rask.jobs.wait(created.id);

const { data: outputs } = await rask.exports.listProjectOutputs(processed.result.project_id);
console.log(outputs.data.find((o) => o.kind === 'dubbed_video')?.url);
API=https://api.rask.ai/v3
AUTH="Authorization: Bearer $RASK_API_KEY"
ORDER="{\"media_id\":\"$MEDIA\",\"target_language\":\"es-es\"}"

# the price first - free, and it changes nothing
CAP=$(curl -s -X POST $API/projects/estimate -H "$AUTH" \
  -H "Content-Type: application/json" -d "$ORDER" | jq -r .total_minutes)

# the price you accepted is the ceiling
PROJECT=$(curl -s -X POST $API/projects -H "$AUTH" \
  -H "Content-Type: application/json" -H "Idempotency-Key: $(uuidgen)" \
  -d "$(jq -c --argjson cap "$CAP" '. + {expected_cost_minutes:$cap}' <<<"$ORDER")" | jq -r .project_id)

curl -s $API/projects/$PROJECT/outputs -H "$AUTH" \
  | jq -r '.data[] | select(.kind == "dubbed_video") | .url'
# the same work, asked for instead of written
claude mcp add --transport http rask https://api.rask.ai/v3/mcp
claude mcp login rask

> Dub this video into Spanish and keep my voice.

  Rask - that will use 4 minutes of your balance. Approve?
  Done. The Spanish version is saved next to the original.
FAQ

Rask MCP server: common questions

Still deciding? Talk to our team.

What is the Rask MCP server?

Which AI assistants does it work with?

Is Rask MCP a local or a remote MCP server?

How do I connect it?

Do I need an API key?

How is this different from the Rask API?

How much does it cost?

Which plans include it?

How do I keep an assistant from spending too much?

Can it work with files on my computer?

Can it clone a voice?

Can it translate the text written inside the video?

Dubbing takes minutes. What happens while a job runs?

Which languages can it dub into?

How do I disconnect it?

Does it work behind a company firewall?

What happens to my data?

Ask your assistant for the next language

Connect Rask once, then describe what you need. The finished video arrives in the same conversation.