Feature Scan
Scan any Jira issue to see the complete test health picture for that feature: linked test cases, automation and manual results, coverage gaps, and an overall risk score - all aggregated in one report.
Prerequisites
Feature Scan is built on top of the Jira integration. Connect Jira under Settings → Integrations first - the scan search, issue hierarchy, and ticket metadata all come from your Jira site. See Integrations for setup steps.
Scanning a Feature
Search for a Jira ticket in the scan bar and select it. Any issue type works - epics, stories, tasks, and bugs. When you scan an epic, all child issues (and their children, two levels deep) are aggregated into a single report, so you get feature-level totals rather than a single ticket's view.
The scan discovers test artifacts from two directions:
- Direct links - test cases linked to the scanned issue or any of its child issues.
- Via executions - cases that belong to a manual execution linked to the ticket. These are tagged "via execution" in the case table so you can tell the provenance apart.
Linked autotests are then resolved through case-to-autotest links, which powers the automation metrics and run history.
Jira Context
The top card shows the scanned ticket exactly as Jira sees it: key, issue type, workflow status, priority, assignee, and summary, plus an expandable description. For epics, an expandable Child Issues tree lists every child with its own status and how many cases and autotests are linked to it. Children with nothing linked are flagged "no tests linked" - an instant map of where coverage is missing inside the feature.
Coverage Metrics
Six stat cards summarize the feature's test health. Values are colored by health bands (good / warning / critical) so problem areas stand out.
Total linked cases, split into automated and manual-only counts.
The share of linked cases that have at least one recorded result, automated or manual.
Pass rate across the latest automated results for the linked cases.
Pass rate across manual execution results for the linked cases.
Cases with no results at all - the most direct coverage gap signal.
A low / medium / high verdict derived from the metrics above. The risk signals that triggered it (for example failing autotests or untested cases) are listed as chips below the cards.
Linked Test Cases
A table lists every discovered case with its ID, title, priority, latest automated result, latest manual result, and the Jira key it is linked through. Expand a row to see the case's recent history: the last automation launches and manual executions it appeared in, each with status and date. Every entry links through to the launch, execution, or case detail page.
Launches, Executions & Releases
Below the case table, the scan lists the feature's activity across the rest of the platform:
- Automation Launches - runs that include the feature's linked autotests, with per-launch pass/fail/skip counts and the matching tests highlighted.
- Manual Executions - executions containing the linked cases, with per-case statuses and who executed them.
- Releases - releases connected to the ticket, with their linked launch and execution counts and date ranges.
Watchlist
Star a scanned feature to pin it to your watchlist. Watchlist cards show the feature's key metrics - case count, coverage, automated and manual pass rates, risk score - along with when it was last scanned. Metrics refresh automatically in the background, and clicking a card re-runs the scan, making the watchlist a living dashboard for the features you care about most.
Feature Scan via MCP
The same report is available to AI assistants through the get_feature_scan MCP tool, so you can ask Claude or Cursor "how well is PROJ-123 tested?" and get the full coverage breakdown. See MCP Tools for details.
link_autotest_to_case MCP tool) to light up the automation metrics.