a GitHub App built with probot that sends out periodic reports
Disclaimer: This report Bot is heavily focused on our setup, so in order to use it you probably need to fork it or help us make it more general purpose.
The bot is activated by the file .github/report.yml
in the settings
repository. If not configured otherwise, this is the probot-settings
repository in your organization.
The file can be empty, or it can override any of these default settings:
# Local times at which the bot will send automatic reports
reportTimes:
- 09:00
- 12:30
# Timezone offset for all users where the timezone cannot be inferred
# Defaults to PDT (UTC-07:00)
defaultTimezone: -420
# Maximum number of days to report new issues
newIssueDays: 7
# Ignores all issues that match this regular expression in their title
ignoreRegex: "\\bwip\\b"
# Ignores all issues with these lables
ignoreLabels:
- duplicate
- wontfix
- invalid
# Mailer configuration, can be omitted to disable email
email:
# Name of the email sender
sender: '"🤖 Eos - Github Bot" <noreply@md.getsentry.com>'
# E-Mail subject
subject: "Github needs your attention"
# Template for the entire email body
bodyTemplate: >
Hi <%- user.name %>,
<%= toReview %>
<%= toComplete %>
<%= newIssues %>
# Template to render a single issue
issueTemplate: >
<% _.forEach(issues, function (issue) { %>
<li>
<b><a href="<%= issue.html_url %>">
<%- issue.repository_url.match('[^/]+/[^/]+$')[0] %>#<%- issue.number %>
</a></b>:
<%- issue.title %><br />
<small>
opened <%- moment(issue.created_at).fromNow() %>,
updated <%- moment(issue.updated_at).fromNow() %>
by <a href="<%= issue.user.html_url %>"><%- issue.user.login %></a>
</small>
</li>
<% }) %>
# Template to format the section of PRs to review
toReviewTemplate: >
<p>These pull requests need to be <b>reviewed</b>:</p>
<ul>
<%= issues %>
</ul>
# Template to format the section of PRs to complete
toCompleteTemplate: >
<p>These pull requests need to be <b>handled</b>:</p>
<ul>
<%= issues %>
</ul>
# Template to format the section of new issues
newIssuesTemplate: >
<p>There are issues you could <b>label and assign</b>:</p>
<ul>
<%= issues %>
</ul>
This Probot app requires authentication tokens and credentials for third party
apps in environment variables. The project contains a template for environment
variables located at .env.example
. Copy this file to .env
in the project
root and adjust all environment variables.
First, create a GitHub App by following the instructions
here. Then,
make sure to download the private key and place it in the root directory of this
application or set it via the PRIVATE_KEY
environment variable. Finally, set
the following environment variables:
Name | Description |
---|---|
APP_ID |
Unique ID of the GitHub App |
WEBHOOK_SECRET |
Random webhook secret configured during app creation |
SETTINGS_REPO |
optional. Repository to store configs in. |
Within your organization, create a repository to store the configuration file.
If not configured otherwise, it defaults to probot-settings
.
The bot can send report emails to all organization members with configured email
addresses (defaulting to their public email address). This requires a
Sendgrid account. Once created, configure the API token
as SENDGRID_TOKEN
environment variable.
Leave this value empty to skip report emails.
The bot can connect to a Slack team and send summaries there. To do so, it needs
to be registered as Slack bot. Once it has been created, configure its token in
the SLACK_TOKEN
environment variable.
Leave this value empty to skip connection to Slack.
To start the development server, make sure the following environment variables are set:
Name | Description |
---|---|
DRY_RUN |
Disables actual releases. Set to true |
SUBDOMAIN |
Subdomain for localtunnel to receive webhooks |
LOG_LEVEL |
Sets the loggers output verbosity. Set to debug |
Then, install dependencies and run the bot with:
# Install dependencies
yarn
# Run the bot
yarn start
# Run test watchers
yarn test:watch
We highly recommend to use VSCode and install the recommended extensions. They will configure your IDE to match the coding style, invoke auto formatters every time you save and run tests in the background for you. No need to run the watchers manually.
The bot includes an automated test suite that includes unit tests, linting and
formating checks. Additionally, this command generates a coverage report in
coverage/
. You can run it with npm:
yarn test
We use prettier for auto-formatting and eslint as linter. Both tools can automatically fix most issues for you. To invoke them, simply run:
yarn fix
If you would like to run your own instance of this app, see the docs for deployment.
This app requires these Permissions for the GitHub App:
- Repository contents: Read & write
- Organization members: Read-only
Also, the following Events need to be subscribed:
- Push: Git push to a repository
- Membership: Team membership added or removed
Also, make sure all required environment variables are present in the production environment.