Get Started

Video in. Context out.

Everything your agent needs to understand videos.100% open source

What is video2ctx?

video2ctx is an open-source toolkit for connecting AI agents to video data. Give it a YouTube URL and it returns transcripts, frames, comments, playlists, channel details, and metadata as structured context, with timestamps and links back to the source.

Start with the skills for direct work. Pair them with the @video2ctx/cli for authenticated hosted operations, or use the all-things-youtube npm package from server-side TypeScript.

What comes back

  • Transcriptevery line, and the second it was said
  • Channelwho published it, as a resolvable identity
  • Commentsthe audience underneath, pinned state intact
  • Playlistthe series it sits in, in order
  • Searchthe way in when you have a question, not a link

Install once. Pick the right route.

YouTube Direct needs only the skill. Hosted agent work pairs the CLI with the skills. Applications can use the API or npm library.

npx skills add devhims/video2ctx

Choose either or both skills. youtube-ctx works immediately without an account or hosted service.

Choose your route.

Each surface has a clear job, whether an agent, application, or person is doing the work.

  • YouTube Direct

    Install the skills and agents can handle one-off public YouTube requests directly. No CLI, account, API key, or hosted service is required.

    Browse the skills

  • CLI + Skill

    Install both for hosted agent work. The CLI handles browser authentication and stable commands; the skills provide routing and workflow guidance.

    @video2ctx/cli

  • Hosted API

    Normalised JSON with the source links intact. Bearer auth, versioned under /v1.

    Read the reference

  • NPM Package

    A server-side TypeScript client for YouTube data, with no hosted service in the path.

    all-things-youtube

  • Workspace

    Collect moments into projects, ask questions across them, and monitor channels over time.

    Open the dashboard

Start small. Pay when you build.

Every plan includes the dashboard and API. Credits are charged according to the data you request.

Starter

FreeNo card required

1,000 credits once

Explore complete video data in the dashboard and API before committing.

  • Dashboard and API access
  • Transcripts, comments, channels, and playlists
  • Projects and monitoring
  • One credit grant per account
Start free

Builder

Recommended
$20per month

20,000 credits monthly

Build production applications, agents, and internal tools with recurring usage.

  • Everything in Starter
  • Monthly credit renewal
  • Higher project, monitor, and import limits
  • Standard support
Start building

Scale

CustomBuilt around your workload

Custom credit allocation

Run video2ctx inside a product or data pipeline with company-level requirements.

  • Volume pricing and custom limits
  • Invoice billing options
  • Priority onboarding and support
  • Security and SLA options

Tell us what you need at scale.

Share a few details so we can respond with useful limits, pricing, and onboarding options.

By submitting, you agree that we may contact you about this inquiry. See our privacy policy.

Context you can check.

Every moment keeps its source
A transcript segment says what was said and when. Each one links back to that second in the video, so you can play it and hear it for yourself.
Missing data says so
Plenty of videos have comments switched off or no captions at all. You are told that plainly, rather than getting an empty result you have to interpret.
Nothing is locked in
Apache 2.0, and the workspace and API are one codebase. Use the hosted version, install the package, or run all of it yourself.
YouTube first, not YouTube only
Provider routes are namespaced, so a second source slots in beside YouTube instead of changing the shape of what you already handle.

Built to be cited, not scraped.

Every segment keeps the timestamp it came from, so an agent can point at the source instead of paraphrasing it.

Open the workspaceGet an API key