kbme

Gather kanban metrics from your JIRA instance:

• Lead time
• Cycle time
• Throughput

This tool was inspired by one of the talks on the european Atlassian Summit 2017

Watch the video to understand how metrics are collected.

Installation

npm install -g kbme

Usage

Default output is prettified json

kbme >> metrics.json

Use the kbme command to gather information for a particular period of time

kbme --csv --start 2017-08-01 --finish 2017-08-31 >> metrics.csv

Collect metrics for the last 90 days using 15 days intervals

kbme --report --interval 15 --period 90 --csv >> metrics.csv

Adds a new line to the previous report using data from last 15 days

kbme --period 15 --csv >> metrics.csv

Options

-c, --csv         Outputs comma separated values [false]
-k, --keys        Comma separated list of projects keys
-t, --types       Issue types filter [NOT IN (Epic)]
-d, --done        Done status [Done]
-t, --todo        ToDo status [ToDo]
-s, --start       Period start [2017-08-01]
-f, --finish      Period finish [2017-08-15]
-q, --query       JIRA JQL Query [project IN (%keys) AND status IN (%done) AND resolutiondate > %start AND resolutiondate < %finish AND issuetype %types]
-u, --user        JIRA username []
-p, --pass        JIRA password
-j, --jira        JIRA REST API URL [https://server-name.com]
-e, --endpoint    JIRA JQL /search endpoint [%jira/jira/rest/api/2/search?jql=%jql&expand=changelog&maxResults=1000]
--period          Period of time to collect metrics, in days [90]
-i, --interval    Interval to split metrics on when using --report, in days [15]

Automatic environment variables detection

You can create a .env file from which kbme will read environment variables, use this to setup a project you want constant metrics from.

Sample .env file

KBME_JIRA="http://server-url.com"
KBME_USER="MrBobry"
KBME_PASS="DzienBobry"
KBME_AUTO=15
KBME_REPORT=false
KBME_PERIOD=365
KBME_INTERVAL=15
KBME_START="2017-01-01"
KBME_FINISH="2017-06-01"
KBME_DONE="Done"
KBME_TODO="In Progress"
KBME_KEYS="ELEMENTS"
KBME_TYPES="NOT IN (Epic)"
KBME_QUERY="project IN (%keys) AND status IN (%done) AND resolutiondate > %start AND resolutiondate < %finish AND issuetype %types"
KBME_ENDPOINT="%jira/jira/rest/api/2/search?jql=%jql&expand=changelog"

Pay special attention to the query and endpoint options, they have special characters (%opt) that can be used as a template to replace by another option value.

Debug

Log JQL queries into the console

DEBUG=jql kbme

Contributing

Feel free to fill an issue and submit a pull request if you find any problems.

IMPORTANT: Always create feature branches from the beta branch.

Automated versioning

We use semantic-release to automate the versioning process, make sure you follow the commit message convention explained here.

HEADS UP: If you are not sure how write a commit message, make your changes in your feature branch and run npm run commit and follow the assistant.

Releases

Beta

Create a feature branch and make a pull-request to beta branch. Once its merged, you can try and install the package using @beta dist tag on npm.

> npm i -g kbme@beta

Production

Create a new pull-request from beta to master branch. Once it gets merged, the final version will be released using @latest dist tag on npm.