Synthesia MCP server for your AI

Ask for a video and get the download link back. Your AI writes to a presenter, starts the render and waits for it, and it tries a free watermarked version first so nothing is spent by accident.

PopMCPSynthesia

Make a free test video welcoming new customers, with a French presenter, and give me the link

Contact sales Connect

Supported Synthesia tools

Every endpoint in Synthesia's API, plus tools that find avatars and voices by name, check template variables before anything starts, handle the upload for dubbing, and wait for the finished video.

42 tools in the full catalog: 23 read and 19 write. 35 load by default, and the rest are found with tool search.

synthesia_create_video

Start a video with the full set of options, returning at once.

synthesia_delete_video

Delete a video.

synthesia_find_stock_avatars

Find a stock avatar's id for synthesia_render_video, by name or gender.

synthesia_find_voices

Find a voice id for synthesia_render_video, by language ("fr", "fr-CA" or "French"), gender (male or female) and name.

synthesia_get_video

Check one video once: status, duration, a time-limited download link, thumbnail and captions when complete.

synthesia_get_video_thumbnail

Get a finished video's thumbnail so you can see it.

synthesia_list_videos

List videos with id, title, status, visibility, duration, dates and the download link once complete.

synthesia_render_video

The easiest way to make a video: give a title and a script, and this starts it and waits for the download link.

synthesia_render_video_from_template

Make a video from a template: checks your variable names against the template, starts it and waits for the download link.

synthesia_update_video

Change a video's title, description, share setting or call to action.

synthesia_wait_for_video

Wait for a video that is still rendering, then get its download link, thumbnail and captions.

synthesia_add_localization_locales

Add more languages to an existing localization.

synthesia_create_localization_from_media_url

Dub a video that sits behind a presigned S3 link into up to 20 languages.

synthesia_dub_video

Dub a video into up to 20 languages from a public link, a presigned S3 link, a file in base64 (up to 100 MB) or an asset_id.

synthesia_get_localization

Check one localization once: status (queued, processing, completed, failed), done and any error.

synthesia_list_localization_languages

List the languages videos can be dubbed from and into.

synthesia_wait_for_localization

Wait for dubbing to finish, then list each language with its status, video id, download link, thumbnail and captions.

synthesia_get_video_translations

Check the translation of a Studio video, one entry per language: complete, in_progress or error with a code.

synthesia_get_video_xliff

Export a video's text as XLIFF 1.2: script, chapters and text elements, to translate elsewhere.

synthesia_list_translation_languages

List the language codes a Studio video can be translated into with synthesia_start_video_translation.

synthesia_start_video_translation

Translate a Studio video into one or more languages.

synthesia_translate_video_from_xliff

Make a translated copy of a video from translated XLIFF.

synthesia_create_webhook

Have Synthesia POST to a link when a video finishes or fails, as an alternative to waiting.

synthesia_delete_webhook

Delete a webhook so its link stops getting events.

synthesia_list_webhooks

List webhooks with id, link, status and dates.

synthesia_list_audit_events

Read who did what and when across a workspace or organization.

synthesia_check_key_access

See which parts of Synthesia this API key can reach: the Video API (Legacy (v2) scope), Localization (read) and the credit balance.

synthesia_create_interactive_avatar

Turn a Studio avatar into an interactive one.

synthesia_get_credit_balance

Get this workspace's credit balance for the current billing period: total, consumed and available (decimal strings), with the period's start and end.

synthesia_get_media_asset

Check an uploaded asset: awaiting_upload, processing, ready or error, with the language detected.

synthesia_get_template

Get one template with the exact variables it takes: id (the name, which is case sensitive), type and any default value.

synthesia_list_templates

List video templates with their ids, titles and the variables each one takes.

synthesia_start_interactive_avatar_session

Start a live interactive avatar session in your own LiveKit room.

synthesia_upload_asset

Upload an image or video to use as a background or scene element.

synthesia_upload_script_audio

Upload your own narration audio to use in place of a written script.

synthesia_create_video_from_template

Start a video from a template, returning at once.

synthesia_list_localization_locales

List each language of a localization with its status, video id, time-limited download link, thumbnail and captions.

synthesia_get_webhook

Get one webhook: link, status and dates.

synthesia_export_audit_events

Export audit events as CSV text, up to 10,000 rows.

synthesia_search_audit_events

The same audit search as synthesia_list_audit_events, for when you filter by more than 50 actions or actors.

synthesia_create_media_asset

Register a video for dubbing and get temporary S3 upload credentials, for files too big for synthesia_dub_video, when you can run an upload yourself.

synthesia_create_dubbing_project

Start dubbing an uploaded video asset.

Connect Synthesia to Claude, ChatGPT and Claude Code

Credentials off one Synthesia screen, then an endpoint your client signs into. Nothing to deploy, and nothing to install.

Paste an API key from Developers, then API keys, in Synthesia. Tick Legacy (v2) for videos and Localization (write) for dubbing when you make it. PopMCP tries each part with free reads and tells you if one is missing.

API key••••••••••••3f7a

ConnectionVerified

Drop the PopMCP URL into Claude, ChatGPT, Claude Code, Codex or any other supported client. It signs in through PopMCP with OAuth, so your Synthesia credential never leaves the server.

app.popmcp.com/mcp

Claude wants to connectApprove

Ask for something harmless before the agent changes anything. Writes run as soon as it calls them, so confirm it reads the right account first.

Make a free test video welcoming new customers, with a French presenter, and give me the link

synthesia_find_stock_avatarsok

Then just ask for the work

“Make a free test video welcoming new customers, with a French presenter, and give me the link”
- find_stock_avatars
- find_voices
- render_video
- wait_for_video
“Make a video from my onboarding template for Sam at Acme”
- list_templates
- get_template
- render_video_from_template
“Dub this product video into French, German and Spanish and give me each download link”
- list_localization_languages
- dub_video
- wait_for_localization
“Translate my latest training video into Japanese”
- list_videos
- list_translation_languages
- start_video_translation
- get_video_translations
“How many credits do I have left, and which videos did I make this week?”
- get_credit_balance
- list_videos
“Use this picture as the background for a short video announcing our sale”
- upload_asset
- render_video

Contact support View docs

Use Synthesia from the assistant you already open

One hosted endpoint, registered once per client. The same tools and the same permissions, whichever window the request comes from.

Claude DesktopCustom connector

Add the PopMCP URL under Customize → Connectors and the tools appear in every chat.

Claude WebCustom connector

The same endpoint, registered in the browser. Nothing to install on the machine you are working from.

ChatGPTCustom connector

CodexDesktop & CLI

Wire the endpoint into Codex or the Codex CLI when a task needs live business context, not a fixture.

OpenCodeCoding agent

ZCodeDesktop app

Add PopMCP as an MCP server in Z.ai's desktop app and work the account from the same window.

Contact support View docs

FAQ

Synthesia MCP server questions

What teams ask before putting Synthesia behind an AI client.

What does the Synthesia MCP server need to connect?

A Synthesia API key. Make one in Synthesia under Developers, then API keys, and paste it into PopMCP. There is no app to review and nothing to deploy. The key is only shown once, so copy it when you make it.

Which Synthesia plan do I need?

Making videos through the API needs a Pro plan or above, because that is what can grant the Legacy (v2) scope. Dubbing through the Localization API is on every plan, as long as the workspace has credit. Audit logs and XLIFF are for Enterprise.

Will my AI spend my Synthesia credit?

Only if you let it. New PopMCP connections start read-only, so nothing can be made until an owner or admin allows write actions. After that the video tools make a free test with a watermark unless told otherwise, and test videos are limited to 30 a day.

How long does a video take?

Usually a few minutes, and a finished video stays in progress while it passes moderation review. Your AI waits for a while, then picks the same video up again, so nothing starts twice. Download links are time-limited, and a fresh read gives a new one.

Does my AI need to know avatar and voice ids?

No. Synthesia's API has no list of them, so PopMCP reads Synthesia's published tables, and your AI can find a presenter by name or a voice by language. A custom avatar's id is under Copy ID on the Avatars page in Synthesia.

Can it dub a video that is just a link?

Yes. Give a public link or a file up to 100 MB, and PopMCP uploads it and starts dubbing. Lip-sync stays off unless you ask, because it costs extra and needs a plan with it. For a bigger file, a presigned S3 link goes straight through.

Do I have to run or host the Synthesia MCP server myself?

No. PopMCP hosts it and gives you a Streamable HTTP MCP URL. There is nothing to deploy, no callback URL to register, and no infrastructure to keep running or patch.

Which AI clients can use this MCP server?

Any client that supports a remote Streamable HTTP MCP server: Claude Desktop, Claude Web, Claude Code, ChatGPT custom connectors, Codex and the Codex CLI, OpenCode and Z.ai's ZCode, as well as editors such as Cursor, Windsurf and Zed. Your client signs in to PopMCP with OAuth; the provider credential never leaves the server.

How do I connect Synthesia to Claude Code?

Connect Synthesia in PopMCP first, then run claude mcp add --transport http popmcp https://app.popmcp.com/mcp in a terminal. Inside Claude Code, run /mcp and sign in to PopMCP in the browser window it opens, ticking Synthesia on the consent screen. The Synthesia tools are then available in that session, and no Synthesia credential is stored on your machine.

Why do only 35 of the 42 tools load by default?

Loading 42 tool definitions into every conversation burns context before the work starts. PopMCP loads a curated 35-tool set covering the everyday workflows and keeps the rest searchable, so your AI pulls a specialist tool only when a task needs it. Every tool stays searchable, and on Scale the owner can load the full catalog for a teammate's connection.

What is the difference between a tool and a trigger?

A tool is something your AI calls, and there are 42 here. A trigger is something PopMCP watches for you: 4 Synthesia events that can start work on the PopMCP side. Triggers never enter your AI client's context, and none of them ask you to paste a callback URL into Synthesia.

Connect Synthesia to your AI in minutes

Launch a hosted MCP endpoint for Synthesia, keep tool access under control, and start testing real workflow prompts right away.

Connect your first MCP