SkipDBSkipDB

About SkipDB

SkipDB is an open, crowdsourced database of skip timestamps — intros, recaps, outros and previews — for movies and TV shows. The data is contributed by the community, reviewed by the community, and published openly under a license that keeps it free forever.

The problem with the alternatives

Skip-timestamp services have existed for years. Some are built into streaming platforms. Others are community projects. Most of them share the same flaw: the data collected from community submissions is kept proprietary by whoever runs the service.

That means if the service shuts down — or decides to start charging for API access, or gets acquired, or simply stops caring — all of the work contributed by thousands of users disappears with it. You had no say. You contributed the data. You don't own it.

SkipDB exists to fix that. The data you submit is published openly under the Open Database License. Anyone can read it, download it, mirror it, or build on it.

Why open data matters

When community-submitted data stays locked in a private database, the community that built it has no leverage. The operator can change the terms, restrict the API, shut it down, or sell it — and contributors have no recourse.

Open data flips that relationship. Every segment in SkipDB is published in a daily public data dump. Mirrors can stay in sync. Forks can spin up independently. Even if this instance disappears, the data survives.

The license also includes a reciprocity clause: any service that builds on SkipDB data must publish its own corresponding data openly too. No free-riding on community contributions while keeping your own database private.

How it compares

FeatureSkipDBTypical alternatives
Source codeOpen — AGPL-3.0Usually closed
Submitted dataOpen — ODbL 1.0 + reciprocityProprietary, no export
MoviesSupportedMost only support TV
Public data dumpDaily export, no account neededBulk access strictly against their terms
Self-hostableFull mirror or fork in minutesNo
Survives shutdownYes — data is always downloadableData lost when service closes
Multi-stream durationTimestamps shift for different encodesRarely handled
Segment typesIntro, recap, outro, previewUsually intro only

What happens if SkipDB shuts down?

Nothing is lost. The daily data dump is published to GitHub Releases and anyone can download it at any time. The code is open-source so anyone can run their own instance. A fork can be seeded from the public dump in a few minutes and pick up where this one left off.

That's the whole point. SkipDB is designed to be replaceable. The data doesn't belong to whoever runs the server — it belongs to the community that contributed it.

How to host a mirror or take over

Full instructions are on the open data page. Short version: clone the repo, spin up a Postgres database, seed it from the public dump with pnpm db:import, and deploy with NEXT_PUBLIC_READ_ONLY=true to disable submissions and keep the instance read-only.

The GitHub repository has everything you need.