The problem
Most regulated software environments require a Change Request before any production deployment. The CR needs a list of every JIRA issue included in the release — what changed, why, and who approved it.
In practice, that means someone opens the JIRA board, opens the git log, and manually cross-references the two. Commit messages contain ticket numbers like PROJ-1234 fix null pointer in payment service. There might be 30 of them across a two-week sprint. You copy each one, look it up, paste the summary into a doc, repeat. It takes 20–30 minutes and is exactly the kind of task where humans make mistakes — missing a ticket, duplicating one, or pulling the wrong summary.
crgenerator automates this entirely.
How it works
Point it at two git refs — commits, tags, or branch names — and it does the rest:
# Between two commits
crgenerator abc1234 def5678
# Between two tags (common for release-to-release)
crgenerator v1.3.0 v1.4.0
# Between a branch and HEAD
crgenerator origin/main HEAD
It walks every commit in that range, extracts all JIRA ticket numbers from the commit messages, calls the JIRA API to fetch the issue details, and prints a formatted summary ready to paste into your CR.
Setup
Install via Homebrew and configure your JIRA credentials once:
brew install crgenerator
echo 'export JIRA_BASE_URL=https://yourorg.atlassian.net' >> ~/.zshrc
echo 'export JIRA_USER_NAME=you@yourorg.com' >> ~/.zshrc
echo 'export JIRA_API_TOKEN=your-api-token' >> ~/.zshrc
source ~/.zshrc
Your JIRA API token is generated from your Atlassian account settings. After that, run it from any git repository.
Why this matters
The manual CR process is one of those tasks that looks simple but accumulates real cost at scale. A team shipping weekly to a regulated environment spends hours per quarter on this. More importantly, a missed ticket in a CR is a compliance gap — the kind that causes audit findings.
Automating it removes both the cost and the risk. The tool is deterministic: same git range, same JIRA state, same output. It’s also fast enough to run as part of a release checklist without slowing anyone down.
The source is on GitHub, MIT-licensed.