# GitHub data API

GitHub page read November 19, 2026.

Get GitHub repositories as structured rows with repository name, description, stars and forks. Read any public GitHub page with the Extract API, or ask Deep Search for every match on GitHub and across the web, kept up to date, without running a scraper.

SpecStream is the live web data engine for teams and agents.

## What you get from GitHub

Each repository comes back as one row with these fields, and each value carries the URL of the GitHub page it was read on and the time it was read.

| Field |
| --- |
| Repository name |
| Description |
| Stars |
| Forks |
| Primary language |
| License |
| Latest release |

## A real example from GitHub

We read one public GitHub page with the Extract API on November 19, 2026. It returned 3,259 characters of the page's text, and these values are printed on it.

| Field | Value | Source |
| --- | --- | --- |
| Repository | pallets/flask | [github.com](https://github.com/pallets/flask) |
| Stars | 75.2k |  |
| Forks | 17.3k |  |
| License | BSD-3-Clause license |  |
| Commits | 5,566 Commits |  |

## Read one page with the Extract API

Send the address of any public GitHub page and get its main text back as Markdown, with its title, its dates and the time it was read. Each page costs $0.0008, and a page with no text is not charged. This call reads the page above.

curl https://api.specstream.ai/v1/extract -H "Authorization: Bearer \<your key\>" -H "Content-Type: application/json" -d '{"urls": \["https://github.com/pallets/flask"\]}'

## Get every match with Deep Search

Write the request in plain words. Deep Search returns every matching repository on GitHub and across the web as rows with the fields above, and a saved request reruns hourly to yearly with each row labeled changed, unchanged or no longer matching. One request we would run: every public Python web framework repository on GitHub with its stars, license and last release.

## Price

- Extract API: $0.80 per 1,000 pages. A page with no text is not charged.
- Deep Search: 5¢, 15¢ or $1 a delivered row, by what each row needs. You approve the most a run can cost before it starts.
- Search API: $5 per 1,000 searches, or $1 per 1,000 on the fast tier.
- To start: $10 of credit on sign-up, no card.

Source URL and read time on every value.

## More code and models data

- [Docker Hub data API](https://specstream.ai/data/sites/docker-hub)
- [Hugging Face data API](https://specstream.ai/data/sites/hugging-face)
- [npm data API](https://specstream.ai/data/sites/npm)
- [NuGet data API](https://specstream.ai/data/sites/nuget)
- [Packagist data API](https://specstream.ai/data/sites/packagist)
- [PyPI data API](https://specstream.ai/data/sites/pypi)
- [RubyGems data API](https://specstream.ai/data/sites/rubygems)
- [SourceForge data API](https://specstream.ai/data/sites/sourceforge)
- [Every site](https://specstream.ai/data/sites)
- [Extract API](https://specstream.ai/extract-api)
- [Deep Search](https://specstream.ai/deep-search)

Free to quote with credit and a link. Copying in bulk needs a [data licence](https://specstream.ai/data/licence).

## Business lists in the United States

- [Energy management companies in the United States](https://specstream.ai/data/businesses/energy-management-companies/united-states)
- [Boat and ship manufacturers in the United States](https://specstream.ai/data/businesses/boat-and-ship-manufacturers/united-states)
- [Vascular medicine practices in the United States](https://specstream.ai/data/businesses/vascular-medicine-practices/united-states)
- [Produce wholesalers in the United States](https://specstream.ai/data/businesses/produce-wholesalers/united-states)
- [Laser cutting companies in the United States](https://specstream.ai/data/businesses/laser-cutting-companies/united-states)
- [Executive search firms in the United States](https://specstream.ai/data/businesses/executive-search-firms/united-states)
