Filter tokens
The lang and is tokens you can type into any list, and how they translate to GitHub search.
Tokens are words in the search box. They are removed from the text before it is matched, and every token must hold for a repository to stay in the list. Several tokens combine, and so does text.
Tokens
| Token | Keeps | Applies in GitHub scope as |
|---|---|---|
lang:NAME, language:NAME | repos whose main language is NAME, case-insensitive | language:NAME |
is:public | public repos | is:public |
is:private | private repos | is:private |
is:fork | forks | fork:only |
is:source | original repos, not forks | fork:false |
is:archived | archived repos | archived:true |
is:active | repos that are not archived | archived:false |
is:local | repos you have cloned | applied to the results |
is:mine | repos you own | applied to the results |
Tokens are case-insensitive. An unknown is: value is treated as ordinary text.
Where they apply
- My repos and Starred apply every token to the cached list.
- GitHub rewrites tokens into GitHub's own qualifiers.
is:localandis:minehave no GitHub equivalent, so they filter the returned results instead.
GitHub qualifiers
In GitHub scope, these qualifiers are also passed through as typed: language:, is:, topic:, stars:, archived:, fork:, user: and org:.
The tokens language: and is: can still be matched against an owner listing. Any other qualifier sends the whole query to GitHub search.
Combining with text
Free text is matched against the repository name. In My repos and Starred, results with free text are ranked by fuzzy match unless you chose a sort with ctrl+s.
Last updated on