Introduction
This article describes how to use the GitHub REST API with GitHub CLI, curl
, or JavaScript. For a quickstart guide, see Quickstart for GitHub REST API.
About requests to the REST API
This section describes the elements that make up an API request:
Every request to the REST API includes an HTTP method and a path. Depending on the REST API endpoint, you might also need to specify request headers, authentication information, query parameters, or body parameters.
The REST API reference documentation describes the HTTP method, path, and parameters for every endpoint. It also displays example requests and responses for each endpoint. For more information, see the REST reference documentation.
HTTP method
The HTTP method of an endpoint defines the type of action it performs on a given resource. Some common HTTP methods are GET
, POST
, DELETE
, and PATCH
. The REST API reference documentation provides the HTTP method for every endpoint.
For example, the HTTP method for the "List repository issues" endpoint is GET
."
Where possible, the GitHub REST API strives to use an appropriate HTTP method for each action.
GET
: Used for retrieving resources.POST
: Used for creating resources.PATCH
: Used for updating properties of resources.PUT
: Used for replacing resources or collections of resources.DELETE
: Used for deleting resources.
Path
Each endpoint has a path. The REST API reference documentation gives the path for every endpoint. For example, the path for the "List repository issues" endpoint is /repos/{owner}/{repo}/issues
.
The curly brackets {}
in a path denote path parameters that you need to specify. Path parameters modify the endpoint path and are required in your request. For example, the path parameters for the "List repository issues" endpoint are {owner}
and {repo}
. To use this path in your API request, replace {repo}
with the name of the repository where you would like to request a list of issues, and replace {owner}
with the name of the account that owns the repository.
Headers
Headers provide extra information about the request and the desired response. Following are some examples of headers that you can use in your requests to the GitHub REST API. For an example of a request that uses headers, see Making a request.
Accept
Most GitHub REST API endpoints specify that you should pass an Accept
header with a value of application/vnd.github+json
. The value of the Accept
header is a media type. For more information about media types, see Media types.
X-GitHub-Api-Version
You should use this header to specify a version of the REST API to use for your request. For more information, see API Versions.
User-Agent
All API requests must include a valid User-Agent
header. The User-Agent
header identifies the user or application that is making the request.
If you use the Octokit.js SDK, the SDK will send a valid User-Agent
header for you. However, GitHub recommends using your GitHub username, or the name of your application, for the User-Agent
header value. This allows GitHub to contact you if there are problems.
The following is an example User-Agent
for an app named Awesome-Octocat-App
:
User-Agent: Awesome-Octocat-App
Requests with no User-Agent
header will be rejected. If you provide an invalid User-Agent
header, you will receive a 403 Forbidden
response.
Media types
You can specify one or more media types by adding them to the Accept
header of your request. For more information about the Accept
header, see Accept
.
Media types specify the format of the data you want to consume from the API. Media types are specific to resources, allowing them to change independently and support formats that other resources don't. The documentation for each GitHub REST API endpoint will describe the media types that it supports. For more information, see the GitHub REST API documentation.
The most common media types supported by the GitHub REST API are application/vnd.github+json
and application/json
.
There are custom media types that you can use with some endpoints. For example, the REST API to manage commits and pull requests support the media types diff
, patch
, and sha
. The media types full
, raw
, text
, or html
are used by some other endpoints.
All custom media types for GitHub look like this: application/vnd.github.PARAM+json
, where PARAM
is the name of the media type. For example, to specify the raw
media type, you would use application/vnd.github.raw+json
.
For an example of a request that uses media types, see Making a request.
Authentication
Many endpoints require authentication or return additional information if you are authenticated. Additionally, you can make more requests per hour when you are authenticated.
To authenticate your request, you will need to provide an authentication token with the required scopes or permissions. There a few different ways to get a token: You can create a personal access token, generate a token with a GitHub App, or use the built-in GITHUB_TOKEN
in a GitHub Actions workflow. For more information, see Authenticating to the REST API.
For an example of a request that uses an authentication token, see Making a request.
Warning
Treat your access token the same way you would treat your passwords or other sensitive credentials. For more information, see Keeping your API credentials secure.
Parameters
Many API methods require or allow you to send additional information in parameters in your request. There are a few different types of parameters: Path parameters, body parameters, and query parameters.
Path parameters
Path parameters modify the endpoint path. These parameters are required in your request. For more information, see Path.
Body parameters
Body parameters allow you to pass additional data to the API. These parameters can be optional or required, depending on the endpoint. For example, a body parameter may allow you to specify an issue title when creating a new issue, or specify certain settings when enabling or disabling a feature. The documentation for each GitHub REST API endpoint will describe the body parameters that it supports. For more information, see the GitHub REST API documentation.
For example, the "Create an issue" endpoint requires that you specify a title for the new issue in your request. It also allows you to optionally specify other information, such as text to put in the issue body, users to assign to the new issue, or labels to apply to the new issue. For an example of a request that uses body parameters, see Making a request.
You must authenticate your request to pass body parameters. For more information, see Authenticating.
Query parameters
Query parameters allow you to control what data is returned for a request. These parameters are usually optional. The documentation for each GitHub REST API endpoint will describe any query parameters that it supports. For more information, see the GitHub REST API documentation.
For example, the "List public events" endpoint returns thirty issues by default. You can use the per_page
query parameter to return two issues instead of 30. You can use the page
query parameter to fetch only the first page of results. For an example of a request that uses query parameters, see Making a request.
Making a request
This section demonstrates how to make a request to the GitHub REST API using JavaScript and Octokit.js. For a more detailed guide, see Scripting with the REST API and JavaScript.
1. Setup
You must install octokit
to use the Octokit.js library shown in the following examples.
- Install
octokit
. For example,npm install octokit
. For other ways to install or loadoctokit
, see the Octokit.js README.
2. Choose an endpoint for your request
-
Choose an endpoint to make a request to. You can explore GitHub's REST API documentation to discover endpoints that you can use to interact with GitHub.
-
Identify the HTTP method and path of the endpoint. You will send these with your request. For more information, see HTTP method and Path.
For example, the "Create an issue" endpoint uses the HTTP method
POST
and the path/repos/{owner}/{repo}/issues
. -
Identify any required path parameters. Required path parameters appear in curly brackets
{}
in the path of the endpoint. Replace each parameter placeholder with the desired value. For more information, see Path.For example, the "Create an issue" endpoint uses the path
/repos/{owner}/{repo}/issues
, and the path parameters are{owner}
and{repo}
. To use this path in your API request, replace{repo}
with the name of the repository where you would like to create a new issue, and replace{owner}
with the name of the account that owns the repository.
3. Create an access token
Create an access token to authenticate your request. You can save your token and use it for multiple requests. Give the token any scopes or permissions that are required to access the endpoint. You will send this token in an Authorization
header with your request. For more information, see Authentication.
4. Make a request with Octokit.js
-
Import
octokit
in your script. For example,import { Octokit } from "octokit";
. For other ways to importoctokit
, see the Octokit.js README. -
Create an instance of
Octokit
with your token. ReplaceYOUR-TOKEN
with your token.JavaScript const octokit = new Octokit({ auth: 'YOUR-TOKEN' });
const octokit = new Octokit({ auth: 'YOUR-TOKEN' });
-
Use
octokit.request
to execute your request.- Send the HTTP method and path as the first argument to the
request
method. For more information, see HTTP method and Path. - Specify all path, query, and body parameters in an object as the second argument to the
request
method. For more information, see Parameters.
In the following example request, the HTTP method is
POST
, the path is/repos/{owner}/{repo}/issues
, the path parameters areowner: "octocat"
andrepo: "Spoon-Knife"
, and the body parameters aretitle: "Created with the REST API"
andbody: "This is a test issue created by the REST API"
.Note
If you are using a fine-grained personal access token, you must replace
octocat/Spoon-Knife
with a repository that you own or that is owned by an organization that you are a member of. Your token must have access to that repository and have read and write permissions for repository issues. For more information, see Managing your personal access tokens.JavaScript await octokit.request("POST /repos/{owner}/{repo}/issues", { owner: "octocat", repo: "Spoon-Knife", title: "Created with the REST API", body: "This is a test issue created by the REST API", });
await octokit.request("POST /repos/{owner}/{repo}/issues", { owner: "octocat", repo: "Spoon-Knife", title: "Created with the REST API", body: "This is a test issue created by the REST API", });
The
request
method automatically passes theAccept: application/vnd.github+json
header. To pass additional headers or a differentAccept
header, add aheaders
property to the object that is passed as a second argument. The value of theheaders
property is an object with the header names as keys and header values as values.For example, the following code will send a
content-type
header with a value oftext/plain
and aX-GitHub-Api-Version
header with a value of2022-11-28
.JavaScript await octokit.request("GET /octocat", { headers: { "content-type": "text/plain", "X-GitHub-Api-Version": "2022-11-28", }, });
await octokit.request("GET /octocat", { headers: { "content-type": "text/plain", "X-GitHub-Api-Version": "2022-11-28", }, });
- Send the HTTP method and path as the first argument to the
Using the response
After you make a request, the API will return the response status code, response headers, and potentially a response body.
About the response code and headers
Every request will return an HTTP status code that indicates the success of the response. For more information about response codes, see the MDN HTTP response status code documentation.
Additionally, the response will include headers that give more details about the response. Headers that start with X-
or x-
are custom to GitHub. For example, the x-ratelimit-remaining
and x-ratelimit-reset
headers tell you how many requests you can make in a time period.
When you make a request with Octokit.js, the request
method returns a promise. If the request was successful, the promise resolves to an object that includes the HTTP status code of the response (status
) and the response headers (headers
). If an error occurs, the promise resolves to an object that includes the HTTP status code of the response (status
) and the response headers (response.headers
).
You can use a try/catch
block to catch an error if it occurs. For example, if the request in the following script is successful, the script will log the status code and the value of the x-ratelimit-remaining
header. If the request was not successful, the script will log the status code, the value of the x-ratelimit-remaining
header, and the error message.
In the following example, replace REPO-OWNER
with the name of the account that owns the repository, and REPO-NAME
with the name of the repository.
try { const result = await octokit.request("GET /repos/{owner}/{repo}/issues", { owner: "REPO-OWNER", repo: "REPO-NAME", per_page: 2, }); console.log(`Success! Status: ${result.status}. Rate limit remaining: ${result.headers["x-ratelimit-remaining"]}`) } catch (error) { console.log(`Error! Status: ${error.status}. Rate limit remaining: ${error.headers["x-ratelimit-remaining"]}. Message: ${error.response.data.message}`) }
try {
const result = await octokit.request("GET /repos/{owner}/{repo}/issues", {
owner: "REPO-OWNER",
repo: "REPO-NAME",
per_page: 2,
});
console.log(`Success! Status: ${result.status}. Rate limit remaining: ${result.headers["x-ratelimit-remaining"]}`)
} catch (error) {
console.log(`Error! Status: ${error.status}. Rate limit remaining: ${error.headers["x-ratelimit-remaining"]}. Message: ${error.response.data.message}`)
}
About the response body
Many endpoints will return a response body. Unless otherwise specified, the response body is in JSON format. Blank fields are included as null
instead of being omitted. All timestamps return in UTC time, ISO 8601 format: YYYY-MM-DDTHH:MM:SSZ
.
Unlike the GraphQL API where you specify what information you want, the REST API typically returns more information than you need. If desired, you can parse the response to pull out specific pieces of information.
For example, you can get the title and author ID of each issue. In the following example, replace REPO-OWNER
with the name of the account that owns the repository, and REPO-NAME
with the name of the repository.
try { const result = await octokit.request("GET /repos/{owner}/{repo}/issues", { owner: "REPO-OWNER", repo: "REPO-NAME", per_page: 2, }); const titleAndAuthor = result.data.map(issue => {title: issue.title, authorID: issue.user.id}) console.log(titleAndAuthor) } catch (error) { console.log(`Error! Status: ${error.status}. Message: ${error.response.data.message}`) }
try {
const result = await octokit.request("GET /repos/{owner}/{repo}/issues", {
owner: "REPO-OWNER",
repo: "REPO-NAME",
per_page: 2,
});
const titleAndAuthor = result.data.map(issue => {title: issue.title, authorID: issue.user.id})
console.log(titleAndAuthor)
} catch (error) {
console.log(`Error! Status: ${error.status}. Message: ${error.response.data.message}`)
}
Detailed versus summary representations
A response can include all attributes for a resource or only a subset of attributes, depending on whether you fetch an individual resource or a list of resources.
- When you fetch an individual resource, like a specific repository, the response will typically include all attributes for that resource. This is the "detailed" representation of the resource.
- When you fetch a list of resources, like a list of multiple repositories, the response will only include a subset of the attributes for each resource. This is the "summary" representation of the resource.
Note that authorization sometimes influences the amount of detail included in a representation.
The reason for this is because some attributes are computationally expensive for the API to provide, so GitHub excludes those attributes from the summary representation. To obtain those attributes, you can fetch the detailed representation.
The documentation provides an example response for each API method. The example response illustrates all attributes that are returned by that method.
Hypermedia
All resources may have one or more *_url
properties linking to other resources. These are meant to provide explicit URLs so that proper API clients don't need to construct URLs on their own. It is highly recommended that API clients use these. Doing so will make future upgrades of the API easier for developers. All URLs are expected to be proper RFC 6570 URI templates.
You can then expand these templates using something like the uri_template gem:
>> tmpl = URITemplate.new('/notifications{?since,all,participating}')
>> tmpl.expand
=> "/notifications"
>> tmpl.expand all: 1
=> "/notifications?all=1"
>> tmpl.expand all: 1, participating: 1
=> "/notifications?all=1&participating=1"
Next steps
This article demonstrated how to list and create issues in a repository. For more practice, try to comment on an issue, edit the title of an issue, or close an issue. For more information, see the "Create an issue comment" endpoint and the "Update an issue" endpoint.
For more information about other endpoints that you can use, see the REST reference documentation.