The Issue Collector is now the work item collector.
It was renamed, not retired.

Went looking for the Jira Issue Collector and found a page about a work item collector instead? Nothing has been taken away from you: Atlassian replaced the word issue with the phrase work item across Jira, and the Cloud page describing the Issue Collector still lives at a URL that reads use-the-issue-collector. The Issue Collector is documented on Cloud and in the Jira Data Center administration guide alike. So this is a straight comparison rather than a migration guide, and the difference between the two is not whether they exist — it is how much of a bug each one carries, and what happens to the ticket afterwards.

Where they are genuinely good

There is no third party in the path, and that is worth a great deal. The Issue Collector is a Jira feature: an administrator configures a feedback form, pastes it into a page, and work items appear in the project with no other product to connect or trust. The Issue Collector is documented for Data Center as well as Cloud, which means a self-hosted Jira behind your own firewall can still collect feedback from a public site — Snag is a hosted service and has no self-hosted build at all. And the Issue Collector takes reports from website visitors who have no Jira account, so nobody is licensing reporters.

Where Snag is different

Snag is built around one question: can this bug be fixed, and can we prove it? The capture is sized for that. One click records a session replay with the console and the network beside it, notes the repro steps, and files a work item in Jira or an issue in GitHub that links back to all of it. Personal data is masked in the browser before anything uploads — inputs are never recorded. Then a sandboxed fix run reproduces the bug in a disposable copy of your repo, writes a repro test that fails before the fix and passes after, and opens a pull request for a person to review. Nothing merges without a human, and the sandbox is destroyed when the run ends. A run that produces no pull request costs half, and the first 5 fixes are free on every plan.

Which one to pick

Use the Issue Collector when what you need is a feedback form that lands in Jira with no vendor in the middle, when your Jira is self-hosted, or when the reports you get are suggestions rather than defects. Use Snag when the reports are bugs, when whoever picks the ticket up needs the replay and the console in front of them, and when the thing you want back is a reviewable pull request with a test attached. They also coexist perfectly well: nothing stops a team running the Issue Collector on a marketing site and Snag on the product.

Snag and Issue Collector, side by side

Every Issue Collector cell links to the page we read it on.

Questions people actually ask

Was the Jira Issue Collector discontinued?

No — it was renamed. Atlassian replaced the word "issue" with "work item" across Jira, so the Cloud documentation now reads "work item collector" while the URL of that same page still says issue collector. The Data Center administration guide still uses the older wording. What did end, on 15 February 2024, was support for Jira Server, and that is the end-of-life most people are half-remembering. If you searched for the issue collector and landed on a page about a work item collector, you found the thing you were looking for.

What does each one put on the ticket?

The collector's documented default is user context in the environment field: the URL, the User-Agent and the screen resolution. Snag attaches a session replay with the console and the network beside it, the repro steps and an annotated screenshot, and links them from the Jira work item, with personal data masked in the browser before anything uploads — inputs are never recorded. Both are enough for a person to start on. The difference shows up when the next reader is a coding agent, which needs the order things happened in before it can reproduce anything.

Can people file a report without a Jira licence?

On both, yes, and this is genuine parity rather than a difference. Atlassian's own documentation says visitors to your website do not require a user account in Jira to use the feedback form. Snag works the same way: filing a bug never needs an account or a licence, and members are unlimited on every plan including the free one. Nobody is buying seats for reporters in either direction.

What happens to a Snag ticket after it is filed?

It sits in Jira or GitHub like any other ticket until triage, or somebody reading the queue, sends it to a fix run. From there the agent reproduces it in a disposable copy of your repo, destroyed when the run ends, and comes back with a pull request carrying the diff, a repro test that fails before the fix and passes after, and the evidence the agent worked from. Nothing merges without a human, so the ticket closes when a person merges. A run that ends without a pull request costs half, and the first 5 fixes are free on every plan.

Where these details come from

Every Issue Collector detail on this page links to the Issue Collector page we read it on, with the date we read it. Something out of date? Email admin@dvcllc.io and we will fix it.