This is a version of GitHub's chat bot, Hubot. Hubot's pretty cool.
coffee me
Charlie will add you to the queue for virtual coffee, or if someone else is
already in the queue, Charlie will match you up and send you each a DM!
@charlie define <term> or @charlie glossary <term>
Charlie will attempt to look the term up in the
18F Procurement Glossary
and report what it finds
@charlie hug me or @charlie hug bomb <number>
Charlie will send you a picture of a hug-offering colleague, or several!
love @someone for the thing (or ❤️ or <3 them)
Charlie will exclaim about even more love, and will post in the #love channel
@charlie opm status
Charlie will check with OPM about whether or not DC offices are open today
@charlie set tock <tock line> or @charlie set tock line <tock line>
Charlie will associate the current channel to the given tock line
@charlie tock or @charlie tock line
Charlie will tell you what tock line is associated with the current channel, if any
giraffe/dag/cat fact
Charlie will tell you something fascinating.
@charlie when is the next holiday?
Charlie will tell you when the next federal holiday is
Plus more! Try sending a DM to Charlie with help for a full list!
The easiest way to test Charlie locally is using Docker. First, create a .env file
(see the .env-sample for reference) that sets, at minimum, the HUBOT_SLACK_TOKEN
variable. (See the configuration section on environment variables below for a list of
all the variables you can set.) Then, run docker-compose up. This will get all of
Charlie's dependencies installed, setup a redis container, hook up Charlie and redis,
and start Charlie up.
If you don't want to use Docker, you can run npm install from the root directory,
set your HUBOT_SLACK_TOKEN environment variable, and then run sh start.sh. This
is a minimum execution; to enable more features, you may need to set associated
environment variables.
18F's Hubot is named Charlie, and is deployed in Cloud.gov.
Charlie is set up with continuous deployment, just merge your code to main and it will get deployed with CircleCI.
We use hubot-slack-github-issues to automatically create Github issues when certain emoji reactions are added to messages in certain channels. This is configured in config/slack-github-issues.json.
The <token> strings in the top part of the config are replaced by environment variables (see below) if
provided. To add a new Github issue rule, add a new entry to the rules array. The properties are
as follows:
| name | purpose |
|---|---|
| reactionName | The name of the emoji in Slack that should trigger this issue, without the colons. For example, if you want :evergreen: to trigger an issue, this value should be evergreen. |
| githubRepository | The name (and only the name) of the repository where the issue should be filed. The repo owner is taken from the top level of the config, in the githubUser property. |
| channelNames | A list of channel names where this reaction will trigger an issue. This lets you limit responses to just certain channels. If this is not included, then Charlie will respond to the reaction in every channel. |
| name | purpose |
|---|---|
| HUBOT_SLACK_TOKEN | Required to connect to Slack. |
| HUBOT_GITHUB_TOKEN | Required for Github integrations. For example, there's a script that will create a Github issue when certain emoji reactions are added to messages. This token must be set for that to work. |
| HUBOT_TOCK_API | Required for Angry and Optimistic Tock bots. |
| HUBOT_TOCK_TOKEN | Required for Angry and Optimistic Tock bots. |
| ANGRY_TOCK_TIMEZONE | The timezone used for Angry Tock notifications for truant users. Defaults to America/New_York if unset. |
| ANGRY_TOCK_FIRST_TIME | The time of day for the first reminder to Tock truants. This is sent at the same time to all truants, based on the timezone in ANGRY_TOCK_TIMEZONE. |
| ANGRY_TOCK_SECOND_TIME | The time of day for the second reminder to Tock truants as well as supervisors. This is sent at the same time to all truants, based on the timezone in ANGRY_TOCK_TIMEZONE. |
| ANGRY_TOCK_REPORT_TO | A comma-delimited list of channels or users to send the second truancy report to. Defaults to #18f-supes. Specify users with @username. |
| HUBOT_TWITTER_ACCESS_TOKEN_SECRET | Used by the twitter stream integration |
| HUBOT_TWITTER_STREAM_ACCESS_TOKEN | Used by the twitter stream integration |
| HUBOT_TWITTER_STREAM_CONSUMER_KEY | Used by the twitter stream integration |
| HUBOT_TWITTER_STREAM_CONSUMER_KEY | Used by the twitter stream integration |
| ADAPTER | This should either be omitted or set to slack |
| HUBOT_SLACK_GITHUB_ISSUES_CONFIG_PATH | Path to the file that configures the script that creates Github issues when emoji reactions are added. Defaults to config/slack-github-issues.json. The default is accurate for Charlie, but if you needed to override it for testing, you certainly can. |
| HUBOT_LOG_LEVEL | Log level. When using Docker, this is set to debug. |
| REDIS_URL | URL to redis, if desired. The URL should be of the form redis://:password@host:port - note that there is not a username before the password in this URL, because redis does not support usernames. |
General information about Hubot can be found here:
Please read the contribution guidelines before submitting a pull request.
This project is in the worldwide public domain. As stated in CONTRIBUTING:
This project is in the public domain within the United States, and copyright and related rights in the work worldwide are waived through the CC0 1.0 Universal public domain dedication.
All contributions to this project will be released under the CC0 dedication. By submitting a pull request, you are agreeing to comply with this waiver of copyright interest.