Play with these samples to learn how to integrate the openssi-websdk into your own website.

For more information on the project surrounding these samples, take a look at our docs.

We've provided some live samples that you can play with without needing to download or build any code. Use these if you're going through these demos for the first time.

• Gov DMV
• IBM HR
• Big Blue Credit Union

Several institutions, including a fictional DMV called Gov DMV and a fictional company called IBM, have begun issuing verifiable credentials. Gov DMV requires customers to show up at a physical location with traditional, paper-based documents to verify their identity before they can be issued a credential. IBM's HR department performs background checks to determine whether their prospective employees are who they claim to be. Both institutions are investing time and resources to ensure the digital credentials they issue are trustworthy.

Big Blue Credit Union has decided to take advantage of these investments to reduce KYC costs when creating new accounts at their bank. They allow new customers to sign up for checking accounts on their website, using credentials issued by institutions such as Gov DMV and IBM. Once customers have an account at BBCU, they can use a credential issued by BBCU to sign in to their online banking portal without the need to type in a password.

You will be able to see this process in action by following the instructions below:

• Part 1: Issuing a driver's license
• Part 2: Issuing an employee badge
• Part 3: Signing up for a bank account
• Part 4: Logging in using your bank account credential

Verifiable credentials aren't just useful for interactions between businesses and consumers. In the demos below, you'll see how end-users can exchange verifiable information with each other.

• Demo 1: Ordering a drink
• Demo 2: Providing multiple forms of ID

The code used to run the samples you played with above is included, in its entirety, in this repository. You can quickly build your own sample websites from the templates we've provided.

1. Visit your agent account management page and provision three new agents, govdmv, ibmhr, and bbcu. Make sure the box marking the new agents as issuers is checked.

2. Find your account url and record the passwords for the three agents that you created. You can find this information by visiting your account dashboard, clicking on each agent, clicking the Add Device on the General page for the agent, and clicking on Manual Entry in the Register Device popup panel.

3. Make sure all of your agents are capable of issuing credentials. You can determine this via one of two methods:

   1. Open the account dashboard, click on the agent, and check the Agent Role under the General page.

      Agent Name: govdmv
      Agent DID: WpAsRjUvWNdJhgcpcir1TL
      Agent Role: Issuer
      

   2. Use curl to check the agent's role, making sure it is set to TRUST_ANCHOR. Use this curl command to check the agent's role:

      curl -u <agent_name>:<agent_password> <account_url>/api/v1/info
      

4. If any of the agents are not issuers, use the following curl command to update their role:

   curl -u <account_admin_agent_name>:<account_admin_agent_password> -X PATCH \
       <account_url>/api/v1/identities/<agent_name> \
       -H 'Content-Type: application/json' \
       -d '{ "role": "TRUST_ANCHOR" }'
   

5. Install...

   • the Verify Creds mobile app. OR
   • the Verify Creds browser extension.

Build all the sample apps as docker images using the following command:

docker-compose build

1. Setup your .env file with the necessary parameters to connect. The account url value can be found in your welcome email or by visiting your account dashboard, clicking on an agent, clicking the Add Device on the General page for the agent, and clicking on Manual Entry in the Register Device popup panel.

   $ cp .env_template .env
   
   # edit your .env file
   
   $ cat .env
   ACCOUNT_URL=https://my-account.example.com
   
   DMV_AGENT_NAME=govdmv
   DMV_AGENT_PASSWORD=****
   
   IBMHR_AGENT_NAME=ibmhr
   IBMHR_AGENT_PASSWORD=****
   
   BBCU_AGENT_NAME=bbcu
   BBCU_AGENT_PASSWORD=****
   
   COUCHDB_USER_NAME=****
   COUCHDB_USER_PASSWORD=****
   

   Note: If you have previously run these samples, the COUCHDB_xxx values need to reflect those of your current couch database. The default values used to be admin/password. If you can no longer remember the credentials for your couch database, you can always remove the couchdb directory and restart the container using docker-compose build couchdb && docker-compose up -d couchdb.

2. Start the issuers.

   docker-compose up -d
   

3. Browse to the localhost urls for the various issuers.

   • Gov DMV
   • IBM HR
   • BBCU

4. Read the setup documentation for each app.

   • Gov DMV
   • IBM HR
   • BBCU

5. Browse to the CouchDB UI to see what the apps are writing to the database.

If you don't see your apps running or they don't appear to be functioning properly, you can view the logs for the apps to obtain more information:

$ docker ps
CONTAINER ID        IMAGE               COMMAND                  CREATED             STATUS              PORTS                                        NAMES
5e878312417f        bbcu              "/bin/sh -c 'npm sta…"   5 seconds ago       Up 3 seconds        0.0.0.0:8092->3000/tcp                       verify-creds-samples_bbcu_1
8b82caa94b2f        gov-dmv             "/bin/sh -c 'npm sta…"   4 minutes ago       Up 3 seconds        0.0.0.0:8090->3000/tcp                       verify-creds-samples_dmv_1
2b97afa23e47        ibm-hr              "/bin/sh -c 'npm sta…"   4 minutes ago       Up 3 seconds        0.0.0.0:8091->3000/tcp                       verify-creds-samples_ibm-hr_1
63fd208de010        couchdb             "tini -- /docker-ent…"   10 minutes ago      Up 10 minutes       4369/tcp, 9100/tcp, 0.0.0.0:5984->5984/tcp   verify-creds-samples_couchdb_1

$ docker logs verify-creds-samples_bbcu_1 

> [email protected] start /opt/app
> node ./bin/www.js

2019-03-13T12:45:45.669Z [bin/www.js] info: Connecting to couchdb.  Attempt 1 out of 10
2019-03-13T12:45:45.922Z [bin/www.js] info: Connected to couchdb: "{\"couchdb\":\"Welcome\",\"version\":\"2.3.0\",\"git_sha\":\"07ea0c7\",\"uuid\":\"0c06c8cdfd26a14f8aa9f0cad15e46a4\",\"features\":[\"pluggable-storage-engines\",\"scheduler\"],\"vendor\":{\"name\":\"The Apache Software Foundation\"}}\n"
2019-03-13T12:45:45.929Z [app.js] debug: Setting up express app
2019-03-13T12:45:45.938Z [app.js] info: Initializing agent
2019-03-13T12:45:45.979Z [app.js] info: Attempting to create database bbcu_db
2019-03-13T12:45:46.001Z [bin/www.js] info: Listening on port 3000
2019-03-13T12:45:46.476Z [app.js] info: Created database bbcu_db
2019-03-13T12:45:46.477Z [libs/users.js] info: Publishing Users design doc
2019-03-13T12:45:46.586Z [libs/users.js] debug: Users design doc published. rev: 1-7ba3f412031b58ad8a8bd6c6fa96c40d
...

• Test Holder instructions

There are several different parameters that are used to make the sample apps do what they do. You'll probably need to add, remove, or tweak these values in order to transform the samples into your own proof-of-concept application. Here's a complete list of the existing configuration parameters:

• DB_CONNECTION_STRING: The Couchdb service endpoint that the sample app will use to store user records. http://couchdb:5984 in the Docker Compose file is what allows the samples to use the couchdb container in the Docker Compose environment.
• DB_USERS: The name of the Couchdb database where user records will be stored. If the database is not present, the app will attempt to create it at startup. ex. dmv_db
• ACCOUNT_URL: The URL that is assigned to an account on our Public Agency and associated with a single IBMid. The issuer agent should be registered under this account url. ex. https://<account_uuid>.staging-cloud-agents.us-east.containers.appdomain.cloud/
• AGENT_NAME: The name of the issuer agent on the Public Agency account. ex. dmv
• AGENT_PASSWORD: The password associate with the issuer agent.
• FRIENDLY_NAME: The friendly name to attach to connection offers, credential offers, verification requests, etc. If not provided, the issuer's agent name will be used. ex. Big Blue Credit Union
• AGENT_LOG_LEVEL: The log level to set for the openssi-websdk. Defaults to info.
• AGENT_ADMIN_NAME: The agent name for the first agent on your Public Agency account. These agent credentials are used to create the issuer agent if it doesn't already exist. Due to performance issues with creating agents, using these parameters is not recommended or supported.
• AGENT_ADMIN_PASSWORD: The password for the admin agent.
• CARD_IMAGE_RENDERING: The type of rendering that should be used for credentials. Credential rendering only comes into play when the issuer's credential schema has card_front and/or card_back attributes. The available options are described below:
  • none: No credential rendering is performed and the image attributes are left blank on issued credentials. You should stick to this option while in development to keep log messages to a reasonable size.
  • static: Static images are used to fill in the card_front and card_back attributes when issuing credentials. This mode is useful when you don't yet have a true credential rendering service and want to issue credentials with placeholder images. If this mode is selected, there are additional configuration parameters that must be set:
    • STATIC_CARD_FRONT_IMAGE: A path to an image file to be used for the card_front attribute when issuing credentials. This image should be small (<= 4KB) for performance reasons.
    • STATIC_CARD_BACK_IMAGE: A path to an image file to be used for the card_back attribute when issuing credentials. This image should be small (<= 4KB) for performance reasons.
  • branding_server: Credential images will be rendered by a remote service. The service used by the hosted samples apps is not currently exposed to the public, but you could study the inputs to that service from the sample code and build a service of your own. If this mode is selected, you have to provide information about the service and credential templates that should be used:
    • BRANDING_SERVER_ENDPOINT: A URL to POST credential attributes to in order to receive rendered credential images.
    • BRANDING_SERVER_FRONT_TEMPLATE: The template to reference when asking the branding service for card_front images.
    • BRANDING_SERVER_BACK_TEMPLATE: The template to reference when asking the branding service for card_back images.
• MY_URL: The public URL for the app. Not currently required or used.
• CONNECTION_IMAGE_PROVIDER: The method for attaching images to connection requests. These connection images are most useful when used in conjunction with the FRIENDLY_NAME to help users identify the source of incoming connection offers. The available options are:
  • none: No images will be attached to connection offers. This is the mode that should be used when developing, in order to keep log messages to a manageable size.
  • static: A static image will be attached to any connection offers. Additional required parameters for this mode include:
    • CONNECTION_ICON_PATH: The path to an image file. This image should be (<= 4KB) for performance reasons.
• SESSION_SECRET: The secret to use when creating and managing sessions for the sample app.
• LOGIN_PROOF_PROVIDER: The method for building verification requests for the verifiable credential login functionality. The available options are:
  • none: Users will not be able to log in using verifiable credentials.
  • file: A proof request described in a file will be used to permit users to log in to their accounts. Required parameters for this mode include:
    • LOGIN_PROOF_PATH: The path to a file describing a login proof request. See the example files used by the samples.
• SIGNUP_PROOF_PROVIDER: The method for verifying credentials when a user attempts to sign up for an account. The options are as follows:
  • none: Users will not be able to sign up for accounts.
  • account: Users will be able to sign up for an account using a driver's license and proof of employment. To modify the signup behavior to your use case, you'll have to write a provider of your own. This mode requires the following additional parameters:
    • SIGNUP_ACCOUNT_PROOF_PATH: The path to a JSON file describing a signup proof request. This file should describe attributes from a driver's license and attributes from an employment badge.
    • SIGNUP_DMV_ISSUER_AGENT: The agent name or agent URL for the DMV issuer. The sample will attempt to establish a connection to this agent in order to acquire a list of published driver's license schemas. Each of these schemas will be added to the restriction list for each driver's license attribute in the proof request.
    • SIGNUP_HR_ISSUER_AGENT: The agent name of agent URL for the HR issuer. Serves the same purpose as SIGNUP_DMV_ISSUER_AGENT.
• SCHEMA_TEMPLATE_PATH: The path to a JSON file describing the credential schema for the issuer. This parameter is configured in the Docker image file for each sample issuer and describes the locations of the driver's license, employment badge, and bank account schema files.
• ACCEPT_INCOMING_CONNECTIONS: A toggle that causes the sample app to pole for incoming connection offers and accept them. This is the mechanism that allows BBCU to connect to Gov DMV and IBM HR in order to get a list of credential schemas when SIGNUP_PROOF_PROVIDER === 'account'.
• ADMIN_API_USERNAME: The username to use to protect the admin UI/API. If this and ADMIN_API_PASSWORD are left blank, the admin panel will not be protected by authentication.
• ADMIN_API_PASSWORD: The password to use to protect the admin UI/API.