Where does the repository data come from?

From GitHub's public REST API, read live. Each figure Zeplik returns, the stars, forks, open issues, latest release, and last push, is GitHub's own current record for the repo you name.

Can it look up a private repository?

No. It reads public repositories only. If a repo is private or does not exist, Zeplik says the lookup found nothing rather than inventing figures.

How fresh is the star count?

It is read live at the moment you ask, and the answer states the time it was read. Counts move constantly, so Zeplik reports them as current as of that read.

Does a GitHub lookup cost credits?

The lookup itself is free. You pay only the normal model-token cost of the answer Zeplik writes around it, shown before you send, exactly like any other message.

Look up a GitHub repo's live stars and releases

Ask about a public GitHub repository and Zeplik reads the real record from the GitHub API, then answers with the actual current star count, forks, open issues, primary language, license, latest release, and last push, as of the moment it read them. A language model has no live GitHub data, so a star count it recalls is stale by potentially tens of thousands and a 'latest' release it names may be several versions old. This reads the live record and, when a repo cannot be found or GitHub is rate limited, says so instead of inventing a figure.

Look up an npm package

The lookup is free; you pay only the normal cost of the answer, shown before you send.

How it works

Name the repository
Ask in plain language, like 'how many stars does vercel/next.js have?' or paste a GitHub URL. Owner/repo or a github.com link both work.
Zeplik reads the GitHub API
It reads the live repository record from GitHub's public API, not the model's memory, and treats the numbers as source data with the time it read them.
Read the live stats
The answer states the current stars, forks, open issues, latest release, and last push, so you compare against the repo's real state, not a stale snapshot.

Worked examples

Real prompts you can open in the composer with the text ready to review. Nothing sends on its own. Swap in your own details.

Prompt: How many stars does vercel/next.js have right now?

Zeplik reads the live GitHub record and returns the current star count with the time it was read, not a number frozen in the model's training.

Open this in Zeplik

Prompt: What is the latest release of facebook/react, and when was it published?

It reports the current published release tag and date from GitHub, so you do not act on a version that is several releases behind.

Open this in Zeplik

Prompt: Show me the open issues, forks, and license for tailwindlabs/tailwindcss

It returns the live counts and license together, so you can gauge activity and terms from the repo's real record.

Open this in Zeplik

What it can and cannot do

The lookup is precise about what it can verify and honest about its edges. Here is the plain posture.

Does itThe real current star count

Zeplik reads the live GitHub record, so stars, forks, and watchers are the counts as of the moment it read them.

Does itThe latest published release

It reads the current release tag and publish date, so a 'latest version' answer reflects what GitHub shows now.

Does itOpen issues, language, and license

Every lookup carries the open-issue count, primary language, and license, so you see the repo's real posture, not a guess.

Does itHonest when a repo is missing

If the repo does not exist or GitHub is rate limited, Zeplik says so rather than inventing a star count to fill the answer.

Does notPrivate or internal repositories

It reads public GitHub records only. A private repo cannot be looked up, and Zeplik will say the lookup found nothing rather than guess.

Does notA judgment on whether a project is good

It reports the figures; it does not certify quality or safety. Use the real stats as inputs to your own evaluation.

Grounded, and honest by default

It cites the figures and their time. Every answer states the counts with the moment they were read from GitHub, so nothing is vague or approximated from memory.
It will not invent a star count. A language model has no live GitHub data, so when a repo is missing or rate limited Zeplik says so rather than manufacturing a plausible number.
Repository data cannot hijack the answer. The GitHub record is treated as untrusted content: Zeplik reads and reports it but never obeys instructions embedded inside a description or release note.

Frequently asked questions

Where does the repository data come from?: From GitHub's public REST API, read live. Each figure Zeplik returns, the stars, forks, open issues, latest release, and last push, is GitHub's own current record for the repo you name.
Why not just ask a chatbot how many stars a repo has?: Because a language model has no live GitHub data, so any star count it recalls is stale by potentially tens of thousands. This tool reads the real current record instead.
Can it look up a private repository?: No. It reads public repositories only. If a repo is private or does not exist, Zeplik says the lookup found nothing rather than inventing figures.
How fresh is the star count?: It is read live at the moment you ask, and the answer states the time it was read. Counts move constantly, so Zeplik reports them as current as of that read.
Does a GitHub lookup cost credits?: The lookup itself is free. You pay only the normal model-token cost of the answer Zeplik writes around it, shown before you send, exactly like any other message.

Related skills

More on Zeplik

Check the repo's real state

Look up a public GitHub repo's real live stars, forks, open issues, and latest release, from the GitHub API, with an honest answer when a repo is not found. Free to start.

Look up a GitHub repo Browse all tools