Getting Started with Benchling Apps
Overview
What are Benchling apps?
Benchling apps provide a new framework for you to customize your teams’ experiences on Benchling with custom applications. It allows you to develop, define (i.e. with a name, description, etc), identify, manage, and permission them independently and with greater granularity than ever before.
What about hosting my app?
We currently do not provide a hosting environment for apps, but highly recommend leveraging AWS’s Lambda and/or EC2 services or the equivalents in other cloud providers. Many customers already leverage these technologies and they can be very simple to get started! They are not required though, and apps can be run anywhere that can connect to Benchling’s APIs, including on-prem servers.
When to use Benchling apps
Benchling apps are intended to be used when building a long-lived application or integration via Benchling’s developer platform. They usually are 1:1 with a project, integrated system, or git repository. EG: “Slack app”, “Jira app”, “Liquid Handler app”. Any time you’re kicking off a new development project, you should likely create a new Benchling app.
They are not strictly necessary when building one-time scripts. For instance, if you want to do a 1-time bulk migration scaling a schema field’s value from “meters” to “centimeters”, creating an app is likely more work than necessary, and you can use your personal credentials to gain access to the APIs.
What can a Benchling app do?
Benchling apps allow you to define an application which has unique credentials and unique permissions (similar to a user) on the Benchling platform. This means each app can be configured, accessed, and permissioned independently.
This can prove useful in a few key ways:
- Developers and Admins can understand which application might have performed a specific action by checking the audit logs or activity logs.
- Users can see when an app took action or created something in Benchling, giving them greater understanding of what’s happening.
- Apps make it easier to follow the principle of least privilege, isolating and limiting the capabilities and access of an app to the minimum necessary to accomplish the specific task.
- Apps have their own individual rate limits, allowing apps to function without impacting other integrations by consuming a shared rate limit.
With the new Beta release of App Canvas, Benchling apps now have access to Benchling-coded UI components with developer-defined inputs and blocks that can handle interaction with the Benchling platform. (see: Introduction to App Canvas)
Getting Started
Getting Access
To get started with creating your first Benchling app, first make sure you have access to the “Developer Platform” capability. Check out how to get access in this Help Center article, or check out our guide on Capability Management more generally.
Where to find apps
Apps can be found in the App Workspace; in Benchling, click "Connections" on the left sidebar and then select "Apps." In the App Workspace, you'll see all the Apps that have been created on your Benchling tenant. Each App has its own homepage where users can view information about the app. The General tab includes information about the app, while the Access tab is where admins of the app can control the app’s access to data in Benchling.
Creating an app
Creating an app can be done from the Developer Console. There are a variety of ways to get to this page, but the easiest is to navigate through your user menu -> Feature Settings -> Developer console (pictured below).
That takes you to the “Developer Console” section of the feature settings page, and opens the “Apps” settings. Clicking the "Create" button provides two options: Creating an app from scratch, or creating an app from an app manifest:
Creating an app from scratch
Selecting "From scratch" allows you to create a Benchling App directly, without defining an app manifest; just provide an app name and a brief description and click "Create":
Once your app is created, you’ll be taken to the app’s settings page, where you can edit the name and description, retrieve the app’s credentials (or delete them).
The app will also appear in the App Workspace:
Clicking on the app will take you back to that app’s settings page.
Creating an app from a manifest
Selecting "From manifest" allows you to create a Benchling App using a manifest file. App manifests are the tenant-independent definition of an app, including all of the app's metadata and configuration details. Check out our App Manifest Reference for more details.
Clicking "From manifest" opens a prompt to upload a manifest file:
When a manifest file is uploaded, a Benchling app is automatically generated from the manifest file and you'll be brought to the app's settings page. Apps created from a manifest differ in some ways from apps created from scratch:
- Apps created from a manifest usually leverage App Configuration & Lifecycle, providing configuration options required for the app to run or to customize how the app functions.
- Apps created from a manifest usually have an "Automatic" lifecycle instead of a "Manual" lifecycle. Apps with an "Automatic" lifecycle are Inactive when created; once the app has been fully configured, installation is completed by "activating" the app. See App Configuration & Lifecycle for more details.
Granting an app access
By default, newly created apps do not belong to any organization, teams, or projects and cannot access any data. To give it access to data on your Benchling tenant, you (or the relevant admin) need to add the app to the various organization(s), team(s), and project(s) as needed.
When giving an app access, make sure to keep in the mind the principle of least privilege and give the app the minimum necessary access to accomplish the desired task.
Adding the app to organizations, teams, and projects is accomplished in the same way as adding users. See the sections below for how to add an app to each.
Authenticating as any user
Apps currently can only act as themselves. If you want to access the APIs as any user on the tenant (including reflecting their permissions) you’ll need to set up OIDC. See our Authentication page for more details.
Adding an app to an organization
Adding an app to an organization will allow the app to access any projects that are accessible to all members of the organization. This is generally necessary when creating a new app. Org admins can add apps in the organization page in the “APPS” tab:
Once added, it will appear in the “APPS” tab for that organization:
Adding an app to a team
Once an app is added to the organization, team or org admins can also add apps to teams, which will similarly give that app access to the projects team members have access to. This can be useful when managing several apps together (“Automation Apps”), or for apps that are specific to a certain team (“Developer Team”):
For instance, I might want several apps to be granted access together under “Automation Apps”, which might be a collection of apps used to automate interactions with systems throughout your organization that all need similar levels of access to projects and data.
Adding the app to the team is similar to adding an app to the org. Search for the app you want to add in the “APPS” tab, and then it will appear in the list of apps on that team. You can similarly remove it from this menu.
Adding apps to projects
Finally, apps must be added to projects with an access policy to actually access the data within. They can be added to projects in the same ways as users, by:
- Adding the organization the app is a member of.
- Adding the team the app is a member of.
- Adding the app directly.
Not an Admin?
If you were unable to add an app to an org or team (for instance, if you are not a tenant, org, or team admin), you can always create an individual project and add your app to it to try it out!
You can add the org, team, or app itself to projects in the “Manage collaborators” section of the project creation menu and the project settings menu.
What happens if an app doesn’t have access?
You can find the project settings menu here on existing projects.
If you’re ever getting errors (such as 403 - Forbidden
or 404 - Not Found
) while interacting with APIs via the app, make sure that the app has access to the relevant project whose object(s) the app is attempting to interact with. As a reminder, an app can carry permissions established at the app, team, or organization level within any given project.
Calling the API as an app
Now that your app has been created and has the access it needs, let’s try calling the API with the app’s credentials! To do this, you must use the OAuth Bearer token flow for apps.
Getting Credentials
The following process is only relevant for apps with a "Manual" lifecycle; apps with an "Automatic" lifecycle receive their credentials via an App Lifecycle Webhook
First, navigate to your app’s settings page to retrieve the client ID:
This ID will not change, and is only an identifier, not a password.
Next, you must retrieve the client secret from the same page:
Here you see that there is no secret for this app yet, and we must generate a new one. As the prompt states when generating the secret, you can only see it once. If you lose your secret, you’ll need to generate a new one, which will delete the existing secret:
Getting a Bearer Token
Now that you have your client ID and client secret, you can use it to call the API to retrieve a Bearer token, allowing you to call the other API endpoints as your app. To do that, you must first call the /token
endpoint. Calling this in cURL
looks like:
curl -X POST "https://{tenant_name}.benchling.com/api/v2/token" \
-H "accept: application/json" \
-H "Content-Type: application/x-www-form-urlencoded" \
-d "client_id={insert_id}&client_secret={insert_secret}&grant_type=client_credentials"
This will return a response that has access_token
(what you need), expires_in
time in seconds, and token_type
:
{
"access_token": "eyJ0eXAiOiJKV1QiLCJhbGciOiJFUzUxMiIsImJuY2giOjF9.eyJzdWIiOiJlbnRfRHJvNEVmZHAiLCJleHAiOjE2MzU0MzkyNDYsImF1ZCI6ImJpb3RlY2h0eC5iZW5jaGxpbmcuY29tIiwiaXNzIjoiYmlvdGVjaHR4LmJlbmNobGluZy5jb20ifQ.ATvA9-7aB61PJGlhR5j3cBGlV6szmnJsAJR7HPAsg-EuLffwbwSMPIEO-FJphXB44O3nRQXOL3z0zUZQYr5glx_eAZjCfqa8tnxCxH496tiI9jNba6KH4ikU5v3D8Ce1-p6TRZNEUPAMtnyZKP-vDCRGgjeKlawKGr7iOZ92mnjvx-gd",
"expires_in": 900,
"token_type": "Bearer"
}
Calling the APIs with a Bearer access_token
access_token
Now you have a token, you can call any of the other endpoints with the Bearer token header, for instance, this is the listEntries
endpoint using cURL
:
curl -X GET "https://{tenant_name}.benchling.com/api/v2/entries?pageSize=5&sort=modifiedAt%3Adesc" \
-H "accept: application/json" \
-H "Authorization: Bearer {access_token}"
Using the SDK
Benchling’s Python SDK will handle Getting a Bearer Token and Calling the APIs with a Bearer access_token for you seamlessly, and is a drop in replacement for the current user API Key Authorization (requires v1.0.2 of the SDK or higher).
Here is how you use client credentials / bearer token in the SDK:
from benchling_sdk.benchling import Benchling
from benchling_sdk.auth.client_credentials_oauth2 import ClientCredentialsOAuth2
auth_method = ClientCredentialsOAuth2(client_id="{id_here}",client_secret="{secret_here}",token_url="https://{tenant_name}.benchling.com/api/v2/token")
benchling = Benchling(url="https://{tenant_name}.benchling.com", auth_method=auth_method)
exampleEntries = benchling.entries.list_entries()
print(next(exampleEntries)) # prints the first page
Compare this to using a user API key with the SDK:
from benchling_sdk.benchling import Benchling
from benchling_sdk.auth.api_key_auth import ApiKeyAuth # User API Key Auth Object
auth_method = ApiKeyAuth("user_api_key") # User API Key
benchling = Benchling(url="https://{tenant_name}.benchling.com", auth_method=auth_method)
exampleEntries = benchling.entries.list_entries()
print(next(exampleEntries)) # prints the first page
As you can see, this is a quick update for those using the SDK!
Using the Interactive Docs
Many of our devs like to use their tenant specific API reference docs (eg: my.benchling.com/api/reference) to help test / construct queries. Right now, those docs also provide a cURL
example of how to make the query. These can still be used with apps, but will a slight modification.
Here is what the cURL normally looks like for the listEntries
endpoint using an API key:
curl -X GET "https://biotechtx.benchling.com/api/v2/entries?pageSize=3&sort=modifiedAt%3Adesc" \
-H "accept: application/json" \
-H "Authorization: Basic {api_key}"
To use an app’s Bearer token (see Getting a Bearer Token section above for retrieving it), you can substitute the “Authorization” header for the Bearer token:
curl -X GET "https://{tenant_name}.benchling.com/api/v2/entries?pageSize=5&sort=modifiedAt%3Adesc" \
-H "accept: application/json" \
-H "Authorization: Bearer {access_token}"
App Features
Seeing app activity and auditing apps
App activity
Once you’ve gone through the Calling the API as an app section, we can look at its activity! We can see when it called the API by the “Last Seen” column in the Tenant, Org, and Team pages under the “APPS” tab:
Auditing apps
Apps are audited alongside users. See example POST curl below for creating a sequence using an access token:
curl -X POST "https://{your-tenant}.benchling.com/api/v2/dna-sequences" \
-H "accept: application/json" \
-H "Content-Type: application/json" \
-H "Authorization: Bearer {access_token}" \
-d "{\"bases\":\"TTAATATACCGGTGCATG\",\"folderId\":\"{destination_folder_API_ID}\",\"isCircular\":true,\"name\":\"MySequence\"}"
I can see that the app created this sequence when I view that object’s audit logs, alongside changes made by users.
What about Events?
Our Events API currently has no explicit link to Benchling apps. We plan to change that in the future, but it isn’t part of the current feature set of Benchling apps. You’ll need to route events from your existing event bus to your app through your own infrastructure. We still recommend that any integrations using events make all their API calls through an app.
Updated 2 months ago
Now that you've got a handle on the basics of Benchling Apps, learn how apps are installed and configured