create_video_from_textLABSCO SUMMARY
cancel_video matters more than it looks when a render takes twenty minutes and costs credits.
A misread brief that runs for twenty minutes is twenty minutes of credits spent on something you will discard. Being able to stop it — and to check the balance before starting — is the difference between an affordable experiment and an expensive one. The telemetry note is worth reading: tool calls are logged to a third-party analytics service with request metadata, which some organisations will want to know about.
What it is
Ngram's endpoint, mirroring their public API one tool per endpoint with typed inputs and outputs. Renders run server-side for roughly 15 to 20 minutes, so the flow is submit and poll.
What you get
- create_video_from_text — submit the render
- get_video_status — poll it
- list_videos — what you have made
- cancel_video — stop one in flight
- get_credits_remaining — the balance
Requirements
Connector OAuth or a bearer key beginning ngs_. Free to start; paid from $23.20 a month billed annually for the equivalent of 3,000 credits a month. Note their privacy documentation: every tool call is logged to PostHog with path, method, user agent, duration and status.
Setup effort
Paste a URL with a key — add the endpoint to your client
USE CASES
01
Rendering a video
Submit it, poll the status, and expect fifteen to twenty minutes.
02
Stopping a bad run
Cancel rather than paying for the full render.
03
Checking affordability first
Read the remaining credits before submitting.
5 tools
Listed in project docs. Your client may show a different list.
get_video_statuslist_videosget_credits_remainingcancel_videoBEFORE YOU INSTALL
What to know first
Tools you get
5 tools named in the project's documentation.
Requires
- an MCP client that can reach a remote endpoint
- connector OAuth or Bearer ngs_
INSTALLATION
Connect your client
Add the endpoint to your client.
Server setup
ENDPOINT URL
https://mcp.ngram.comRequirements
- connector OAuth or Bearer ngs_
