The permissions
section of the manifest.yml
file controls your app's access to
remote resources.
The scopes
list declares which OAuth 2.0 scopes are required by your app
when using the authenticated Product Fetch APIs,
and Product events.
The scopes
list also declares which OAuth 2.0 scopes are required by an app
that uses Forge remote to pass auth tokens to a remote back end.
Define each scope on a new line. Your app should use the minimum set of scopes required. For example:
1 2permissions: scopes: - 'read:confluence-content.summary' - 'write:jira-work'
If your app requires no OAuth 2.0 permissions, you must provide an empty scopes
list as in the example below.
1 2permissions: scopes: []
Forge apps deployed in the development environment always receive all available OAuth 2.0 scopes
Certain platform features, such as the App storage API, are authenticated using OAuth 2.0. For a list of scopes required by these features, refer to Forge scopes.
Product scopes enable a Forge app to request a level of access to an Atlassian product. You can find details about each product operation's required scopes through that product's REST API documentation (specifically, in the operation's Oauth scopes required field). For information about each product event's required scopes, see our events documentation.
For more details about each product's OAuth 2.0 (3LO) and Forge scopes, refer to the pages below:
The content
section declares which Content Security Policy (CSP) options are required by your app
when using custom UI.
The scripts
list declares which sources are allowed for an app's script-src
policy.
In the example below, script-src 'unsafe-hashes'
is included in the CSP header for all modules
using custom UI:
1 2permissions: content: scripts: - 'unsafe-hashes'
Source | Description |
---|---|
unsafe-inline | Allows the use of inline resources, such as inline <script> elements,
javascript: URLs, and inline event handlers. |
unsafe-hashes | Allows the use of specific inline event handlers. |
unsafe-eval | Allows the use of eval() and similar methods for creating code from strings. |
blob: | Allows blob: URIs to be used as a content source. |
<sha-algorithm>-<base64-value> | Allows a specific script to be executed, provided it matches the hash declared here.
The only valid hash algorithms are: sha256 , sha384 , and sha512 . |
The styles
list declares which sources are allowed for an app's style-src
policy.
In the example below, style-src 'unsafe-inline'
is included in the CSP header for all modules
using custom UI:
1 2permissions: content: styles: - 'unsafe-inline'
Source | Description |
---|---|
unsafe-inline | Allows the use of inline resources, such as inline <script> elements,
javascript: URLs, and inline event handlers. |
The external
section declares the external resources that your custom UI app is allowed to access.
In addition, it also covers which external website your Forge function is allowed to communicate with. This covers both Custom UI resolvers and any other Forge functions.
In each section, you can add a list of external domains, which end up as a source in an equivalent CSP directive.
External domains follow CSP protocols and must be in one of the following formats:
An https
or wss
URL, such as https://www.example-dev.com
.
Adding a site URL means
that all resources to this site are allowed – you don't need to add *
at the end.
A valid domain name,
such as www.example-dev.com
A wildcard domain name starting with *
, for example *.example.com
. This includes all
nested subdomains below the specified domain name. Wildcards can be used with subdomains,
for example *.static.example.com
to limit to just the static subdomain.
A generic wildcard to support every domain: *
External domains must not contain any invalid special characters. You can check your domain with the following regex pattern:
1 2^(\*\.)?[.a-zA-Z0-9_\-\/:~#%]+$
The fetch
section has the following configurations:
Configuration | Description |
---|---|
backend | The fetch.backend list declares which external domains your Forge functions can talk to. This applies to both custom UI resolvers and any other Forge functions. |
client | The fetch.client list declares which external sources are allowed for an app's connect-src policy. |
There are two methods for declaring an external URL for your fetch back end:
Directly: this involves listing the domain/s directly in the fetch.backend
or fetch.client
section. When you do, you don't need to specify individual URLs, such as example-dev.com/path
. Adding one domain allows access to any URL on that domain. For example:
1 2permissions: external: fetch: backend: - '*.example-dev.com'
1 2permissions: external: fetch: client: - '*.example-dev.com'
Using a wildcard for a backend
or client
(for example, *.example-dev
) does not include the parent domain. If you need to support both, you need to explicitly add the parent domain as a second entry.
As a remote back end: this involves declaring the URLs in a separate remotes
section, where you can explicitly define its purpose (in this case, as a fetch back end). Upon declaring the remote back end, you can refer to its key in your fetch.backend
or fetch.client
list. For example:
1 2permissions: external: fetch: backend: - remote: remote-backend remotes: - key: remote-backend baseUrl: "https://example-dev.io" operations: - fetch
The fetch
setting in the operations property defines the purpose of the remote back end. You’ll need to do this if you want your app to be eligible for PINNED
status for data residency purposes. See Data residency eligibility for more information.
Calls made to any domain that is not defined in the manifest.yml
file of your app will be
rejected. Learn more about runtime egress permissions.
The fonts
list declares which external sources are allowed for an app's font-src
policy.
In the example below, font-src https://www.example-dev.com/fonts.css
is included in the CSP
header for all modules using custom UI:
1 2permissions: external: fonts: - 'https://www.example-dev.com/fonts.css'
The styles
list declares which external styles are allowed for an app's style-src
policy.
In the example below, style-src https://www.example-dev.com/stylesheet.css
is included in the CSP
header for all modules using custom UI:
1 2permissions: external: styles: - 'https://www.example-dev.com/stylesheet.css'
If you are using specific styles and fonts together in your application, you must include both of them as permissions in your app's manifest file. This is because the styles and fonts are external resources that the app needs to access in order to display correctly. By including these permissions, you are granting your app the ability to access and utilize these resources, ensuring that the text and design elements appear as intended.
The frames
list declares which external sources are allowed for an app's frame-src
policy.
In the example below, frame-src https://www.example-dev.com/embed/page
is included in the CSP
header for all modules using custom UI:
1 2permissions: external: frames: - 'https://www.example-dev.com/embed/page'
The images
list declares which external sources are allowed for an app's img-src
policy.
In the example below, img-src https://www.example-dev.com/image.png
is included in the CSP
header for all modules using custom UI:
1 2permissions: external: images: - 'https://www.example-dev.com/image.png'
The media
list declares which external sources are allowed for an app's media-src
policy.
In the example below, media-src https://www.example-dev.com/video.mp4
is included in the CSP
header for all modules using custom UI:
1 2permissions: external: media: - 'https://www.example-dev.com/video.mp4'
The scripts
list declares which external sources are allowed for an app's script-src
policy.
In the example below, script-src https://www.example-dev.com/script.js
is included in the CSP
header for all modules using custom UI:
1 2permissions: external: scripts: - 'https://www.example-dev.com/script.js'
Rate this page: