/code @devintern/pm DevIntern + Jira
Keep Jira as your source of truth. DevIntern reads issues, posts implementation updates, and opens draft PRs without replacing your board or review workflow.
How it works
What DevIntern does with Jira
Each product path keeps Jira as the handoff surface. No rip-and-replace, no new dashboard.
/pm Draft specs into Jira
- 1 Paste a Figma URL, error log, or rough prompt into @devintern/pm.
- 2 AI scans your repo and drafts a story with acceptance criteria.
- 3 You review, edit gaps, and post the issue to your Jira project.
/code Implement Jira issues end-to-end
- 1 Run devintern against a Jira issue key or JQL batch.
- 2 Agent fetches ADF description, comments, and attachments.
- 3 Creates a branch, implements, self-reviews, opens a PR, and comments back on the issue.
Capabilities
Built for real Jira workflows
Jira Cloud REST API v3 (atlassian.net sites)
ADF descriptions converted to markdown for your agent
Status transitions via workflow status names in settings.json
Feasibility and implementation summaries posted as issue comments
JQL batch runs for draining a filtered backlog
A single run
From Jira ticket to reviewed pull request
A single run takes a Jira issue from key to reviewed pull request. Every stage reads or writes Jira the way your team already uses it.
- 1
Fetch the issue
DevIntern pulls the issue over the Jira Cloud REST API. The ADF description is converted to markdown for your agent, comments from teammates come along (earlier DevIntern comments are filtered out), and attachments are downloaded into the task output directory.
- 2
Feasibility gate
Before writing code, the agent checks whether the issue is implementable as written. A vague issue gets a feasibility assessment comment with concrete questions instead of a confidently wrong pull request.
- 3
Status moves to In Progress
If the issue passes the gate, DevIntern transitions it using the workflow status names you set in settings.json, so the board reflects reality while the agent works.
- 4
Implement and self-review
Your coding agent implements the change on a branch, then reviews and fixes its own diff before any human sees it.
- 5
Deliver back to Jira
With --create-pr, DevIntern opens a pull request, moves the issue to your review status, and posts an implementation summary comment. Incomplete runs move the issue back to To Do with an explanation instead of leaving it stuck.
What gets written back to Jira
Every outcome is written back to the issue, so nobody has to check a separate dashboard. Use --skip-comments to run without comments or transitions.
Feasibility check
Automated Task Feasibility Assessment with the questions blocking implementation
Successful implementation
Implementation summary comment with a link to the pull request
Incomplete run or max turns
Implementation Incomplete comment explaining what remains
Estimation run
Automated Story Points Estimation with the reasoning behind the number
Batch runs
Work through Jira in batches
Anything you can express in JQL becomes a work queue. Pass a single key, several keys, or a query that drains a filtered backlog.
Single issue, PR included
$ devintern PROJ-123 --create-pr Drain the To Do column
$ devintern --query "project = PROJ AND status = 'To Do'" --create-pr Quote project names and statuses that contain spaces, exactly as in Jira issue search.
Just your own tickets
$ devintern --query "assignee = currentUser() AND status = 'To Do'" --create-pr AI story point estimation
$ devintern --estimate --query "project = PROJ AND status = 'To Do'" Writes estimates to your story points custom field and posts the reasoning as a comment.
Jira specifics at a glance
Hosting
Jira Cloud (*.atlassian.net); Server and Data Center are not supported
Query language
JQL via --query for batch selection
Descriptions
ADF converted to markdown for the agent
Status updates
Transitions by workflow status name from settings.json
Estimation
Story points written to the custom field you configure (storyPointsField)
Auth
Atlassian account email plus API token, stored in a local .env
Get connected
Setup guides and configuration
Credentials stay in project-local .env files. DevIntern runs on your machine or server, not a shared implementation cloud.
FAQ
Common questions about Jira
Does DevIntern support Jira Server or Data Center?
No. The integration targets Jira Cloud (*.atlassian.net). Jira Server and Data Center are not supported today.
Can @devintern/code and @devintern/pm both use the same Jira project?
Yes. PM drafts and posts issues; Code picks them up from the same project. Most teams validate with one product first, then run both in the pipeline.
What credentials does Jira need?
A Jira Cloud account email plus an API token with access to the target project. Credentials live in .devintern-code/.env or .devintern-pm/.env on your machine.
Will DevIntern replace our Jira workflow?
No. Status transitions, reviewers, and board structure stay the same. DevIntern adds automation on top of issues you already track in Jira.
What permissions does my Jira user need?
Browse projects and view issues on the target projects, add comments (unless you run with --skip-comments), transition issues if you configure status changes, and edit issues if you use --estimate.
What happens when a status transition fails?
DevIntern logs a warning and continues the run. Typical causes are a status name in settings.json that does not match your workflow exactly, or a transition that is not valid from the issue's current status.
Can I stop DevIntern from writing to issues?
Yes. Run with --skip-comments and DevIntern skips all Jira comments and status transitions for that run while still implementing the change locally.
Try DevIntern with Jira
Connect Jira, run your first workflow, and keep the trackers, repos, and reviewers your team already trusts.
Free for interactive use · FSL source-available · BYO model keys