Blog ·
Jira vulnerability management: findings, due dates, the review
Jira vulnerability management works when every finding that matters becomes one issue with an owner, a severity and a due date, and when the review that feeds those issues happens on a fixed schedule instead of whenever somebody remembers. Jira will not scan anything. Your scanner finds the problems; Jira is where the decision about each one is made, tracked and proved. This guide sets up the project, the fields and the workflow, then covers the part most teams lose first: the weekly triage and the monthly report, which only work if they keep happening.
Statements about Jira's built-in features are as of October 2026, from Atlassian's own documentation.
What Jira does in vulnerability management, and what it does not #
It helps to be precise about the split, because teams that expect Jira to be the scanner end up disappointed, and teams that keep everything in the scanner end up with no record of decisions.
| Job | Where it belongs |
|---|---|
| Finding vulnerabilities in code, images, hosts | Your scanner |
| Scoring a finding (CVSS, exploitability) | Your scanner, copied into Jira |
| Deciding who fixes it and by when | Jira |
| Recording that a risk was accepted, and by whom | Jira |
| Proving the review happened every week | Jira |
| Reporting open findings by age and severity | Jira filters and dashboards |
The scanner knows what is wrong. Jira knows what you decided to do about it. An auditor asks about the second part.
Set up the project #
Use one project for security findings, or one issue type inside each engineering project. A separate project keeps the queue in one place for the security lead and lets you restrict who can see it. An issue type inside each team's project puts the fix next to the team's other work. Small organisations usually do better with the separate project; teams with strong ownership per service often prefer the issue type.
Either way, add these fields:
- Severity: a select list with Critical, High, Medium, Low. Copy it from the scanner; do not re-score by hand.
- CVE or finding ID: a short text field, so a filter can find every issue for one CVE.
- Affected asset: the service, repository or host. A component works if your projects already use them.
- Source: which scanner or report raised it, for when two tools report the same thing.
- Due date: set from the severity, covered below.
Keep the summary readable: CVE-2026-1234 in payments-api (openssl) beats a pasted scanner title.
A workflow with a risk acceptance step #
A vulnerability issue needs one status the usual software workflow does not have.
- Triage: new findings land here. Someone confirms it is real and applies to you.
- To do: confirmed, owned, with a due date.
- In progress: the fix is being made.
- Verify: fixed, waiting for the next scan to confirm it is gone.
- Done: the scanner no longer reports it.
- Risk accepted: a deliberate decision not to fix, for now.
Risk accepted is the step auditors care about most. Make the transition to it require a comment saying why, who approved it, and when it should be looked at again. A finding closed as "won't fix" with no reason is the record that fails a review. Our Jira workflow examples cover how to add a required field on a transition.
"Verify" matters too. Moving a finding to Done because a pull request merged is how the same CVE shows up again in next month's scan. Close it when the scan agrees.
Due dates by severity #
Most teams write a remediation policy that gives each severity a deadline. Your numbers are your own; this is the shape:
| Severity | Typical target | What it means in Jira |
|---|---|---|
| Critical | Days | Due date set at triage, owner paged |
| High | A couple of weeks | Due date set at triage |
| Medium | About a month | Picked up in normal planning |
| Low | A quarter, or accepted | Reviewed in the monthly report |
The due date is what turns the policy into something Jira can report on. Set it at triage, from the severity, every time. If you would rather not set it by hand, our guide to setting the due date automatically in Jira covers Automation rules that do it when the severity field changes.
A filter for overdue findings is then one line of JQL:
project = SEC AND statusCategory != Done AND status != "Risk accepted" AND due < now() ORDER BY Severity DESC
Put that filter on a dashboard where the security lead and the engineering managers both see it. Our Jira dashboard examples show a few layouts.
Getting findings into Jira #
There are three common routes, and most teams use a mix.
- A scanner integration that creates Jira issues directly. Many scanners offer one. It is fast, and it is also the route that floods a project, so filter to High and above, or to findings with a known fix.
- An import. Export the scan as CSV and use Jira's import. Good for a monthly scan of hosts, where the volume is predictable.
- By hand at triage. A person reads the scan report and creates issues for what applies. Slowest, but the issues it creates are the cleanest.
Whichever route you use, someone still has to look at the results on a schedule. Integrations miss findings their filter excluded, imports only run when somebody runs them, and manual triage only happens when it is on someone's list. That is the recurring part of vulnerability management, and it is the part that slips.
The routine that keeps the programme alive #
These are the jobs that come back on a clock. Each one is small, and each one is skipped in exactly the week the team is busiest.
| Routine | Cadence | What goes wrong without a ticket |
|---|---|---|
| Triage new scan findings | Weekly | Findings sit in Triage for a month |
| Review overdue findings with owners | Weekly | Deadlines pass quietly |
| Run and import the host scan | Monthly | The scan runs, nobody imports it |
| Vulnerability report to leadership | Monthly | Written from memory, late |
| Re-review accepted risks | Quarterly | An acceptance from last year is still "temporary" |
| Check scanner coverage (new repos, new hosts) | Quarterly | A new service is never scanned |
The re-review of accepted risks is the one that hurts most when it is missed. An acceptance is a promise to look again, and Jira only reminds you if there is an issue with a date on it.
Make the routine tickets appear on their own #
Clone by hand #
Keep one template issue per routine and clone it each cycle. It costs nothing and works for a security team of one with one weekly triage. It stops working the first week that person is away.
Jira Automation #
A Scheduled trigger can create the weekly triage issue. Atlassian's trigger documentation, as of October 2026, says you can "run the flow at a fixed rate (for example, every 7 days), or use a Cron expression for more complex schedules." Weekly is easy. Monthly and quarterly need a cron expression, and our Jira cron expression guide covers the syntax and the schedules it cannot express. Each run counts toward your plan's automation usage, explained in Jira Automation pricing. The same Atlassian page notes that "Scheduled flows that reach a Failure status for 10 consecutive executions will disable automatically," which is worth knowing for a rule whose whole purpose is to never be missed.
A scheduling app #
Recurring Work Items for Jira is built for this list. You write each routine once in the security project, and the issue appears on schedule:
- Turn the app on for the security project and choose who may schedule.
- Create a rule for the weekly triage, on the weekday the team meets.
- Add the steps as subtasks: pull the new findings, confirm each one, set severity and due date, assign owners.
- Give the title a date with tokens, for example
Vulnerability triage {{now.date}}, so every week has its own issue and the history is searchable. - Add monthly rules for the host scan import and the report, and a quarterly rule for the accepted-risk review.
Rules can skip weekends and holidays, so a triage that falls on a public holiday is not created on a day nobody is there. Every run is written to the rule's history with its outcome, which is useful evidence on its own: it shows the review was scheduled every week. The app does not use your Jira Automation quota, and it is scoped to one project, so the security project's schedule does not touch anyone else's. It is free for up to 10 users.
It creates the routine issues. It does not read your scanner, set severities, or close findings; those stay with your integration and your team.
A starter set of recurring rules #
- Weekly vulnerability triage, every Monday, due the same day. Subtasks: new findings reviewed, duplicates closed, severities and due dates set, owners assigned.
- Weekly overdue review, every Thursday. Description: link the overdue filter, list owners to chase.
- Monthly host scan, first working day of the month. Subtasks: run the scan, import findings, spot-check ten of them.
- Monthly vulnerability report, second working day. Description: open findings by severity, overdue count, accepted risks, trend against last month.
- Quarterly accepted-risk review, first working day of the quarter. Description: link the filter for status "Risk accepted", decide keep, fix or retire for each.
- Quarterly coverage check. Description: compare the scanner's asset list with the list of live services.
If patching is a separate cycle in your organisation, run it alongside this one; our Jira patch management guide covers that monthly cycle, and the Jira risk register guide covers the wider register an accepted vulnerability usually belongs in.
Reports an auditor will ask for #
Most security frameworks and customer questionnaires ask the same few questions. Each one is a saved filter:
- Open findings by severity:
project = SEC AND statusCategory != Done ORDER BY Severity DESC. - Findings fixed late: resolved issues where the resolution date is after the due date. Export and compare, or use a dashboard gadget.
- Accepted risks and their review dates:
project = SEC AND status = "Risk accepted". - Evidence the triage happened:
project = SEC AND summary ~ "Vulnerability triage" ORDER BY created DESC. One issue per week, each with its subtasks done, is a cleaner answer than a meeting invite.
When the manual way is enough #
Not every team needs all of this. If your scanner reports a handful of findings a month and one person owns them reliably, a short list in Jira and a calendar reminder for the triage is enough. Clone the triage ticket each week and move on. If you already run a few Automation rules and stay comfortably inside your plan, a Scheduled rule for the weekly triage is a natural next step. A scheduling app earns its place when the routine list grows past a couple of rules, when monthly and quarterly schedules keep landing on the wrong day, or when you need the history to show an auditor the review never skipped a week.
Frequently asked questions #
Can Jira scan for vulnerabilities? No. Jira tracks the findings and the decisions; a scanner finds the vulnerabilities and many scanners can create Jira issues.
Should every scanner finding become a Jira issue? Usually not. Create issues for findings that apply to you and need action, typically High and above or anything with a known fix, and leave the rest in the scanner's own report.
How do I record a risk acceptance in Jira? Add a "Risk accepted" status, require a comment with the reason and approver on the transition, and schedule a recurring review of everything in that status.
Can Jira create the weekly vulnerability triage ticket automatically? Yes, with a Jira Automation Scheduled rule, or with a scheduling app such as Recurring Work Items for Jira, which creates it on the weekday you choose with its own title, subtasks and due date.