/
/
Prove a Jira fix with a fresh Ghost Inspector run
For developers and QA who want bugs closed on evidence, with the result recorded on the issue, by asking their AI assistant.
On this page
Connector
Connect Jira, pick your AI tool and paste the prompt.
The problem
Bugs get closed on trust and some come back
Before a Jira bug is closed, someone should rerun the Ghost Inspector test it came from and record the result. Finding the test, running it and writing up the outcome is tedious, so bugs close on trust and some return.
Without PopdockAI
- Work out which test the bug came from
- Run it in Ghost Inspector
- Dig out the failing step if it fails
- Close the bug without recording proof
With PopdockAI
- The bug's test is found from its summary, a comment or a name you give
- It stops if several tests match, and asks if the bug is closed
- The test runs once, with the failing step collected on failure
- One verification comment is posted, and the status never changes
The prompt
Use this prompt
Copy it into any AI tool connected to PopdockAI. Swap the values in braces for your own.
**Objective:** Check whether the fix on Jira issue {issue-key} actually holds, by re-running the Ghost Inspector test that the issue was raised from and recording the outcome on the issue. This prompt runs a test and posts one comment. It does not change the issue's status.
**Step 1: Read the issue**
Call `jira-get-issue` on the jira server with `issue-key` set to {issue-key}. If nothing comes back, report that and stop. Keep the summary, status, assignee and project key.
**Step 2: Find the test**
Take the test name from the issue. If the summary begins `Ghost Inspector failure: `, the test name is the remainder. Otherwise, call `jira-get-issue-comment` with `issue-id` set to {issue-key} and look for a comment containing `Ghost Inspector result id: `; take the result id from it and call `ghost-inspector-get-test-result` on the ghost-inspector server with `id` set to that value to read the test name.
If {test-name} was supplied as a parameter, use that instead of either and say in the report that it was overridden.
Call `ghost-inspector-get-test` with `name.contains` set to the test name. If it matches no test, report that the issue cannot be verified automatically and stop. If it matches more than one, list the matches with their id, suite and browser, and stop rather than guessing which to run.
**Step 3: Run it**
Call `ghost-inspector-execute-test` with `test-id` set to the matched test's id. If {browser} was supplied, pass it as `browser`. Leave `immediate` unset so the run completes and the pass state is real.
Runs can take some time. Do not call the tool a second time because the first has not answered yet.
**Step 4: Gather detail if it failed**
If the run did not pass, call `ghost-inspector-get-test-step` with `test-result-id` set to the new run's id and `passing` set to false, and keep the first failing step. Also call `ghost-inspector-get-console-log` with the same `test-result-id` and `error` set to true, `max-records` 10.
**Step 5: Report back onto the issue**
Call `jira-add-issue-comment` with `issue-id` set to {issue-key} and `body` set to a verification note containing: whether the re-run passed; the new Ghost Inspector result id; the browser and execution time; and, when it failed, the failing step and the console errors. Begin the comment with `Ghost Inspector verification run` so later reads can recognise it.
If `jira-add-issue-comment` rejects the issue key, call `jira-get-issue-watcher` with `issue-key` set to {issue-key}, read the issue id it returns, and retry once.
**Step 6: Report**
State the issue key, the test that was run, whether it passed, the new result id, and whether the comment was posted. If it passed, say that the issue is a candidate to close but do not close it.
**Rules & Guidelines:**
* `jira-get-issue`, `jira-get-issue-comment`, `jira-add-issue-comment` and `jira-get-issue-watcher` are on the jira server. `ghost-inspector-get-test`, `ghost-inspector-get-test-result`, `ghost-inspector-execute-test`, `ghost-inspector-get-test-step` and `ghost-inspector-get-console-log` are on the ghost-inspector server. All three servers must be connected.
* The linking convention used by every prompt on this server: a Jira issue raised from a Ghost Inspector failure has a summary beginning `Ghost Inspector failure: ` followed by the test name, and its first comment contains the line `Ghost Inspector result id: ` followed by the result id. Matching in both directions depends on this, so do not vary the wording.
* Do NOT call `jira-update-issue` or `jira-assign-issue`. Moving the issue to Done is a person's decision, and this prompt only supplies the evidence for it.
* Do NOT call `ghost-inspector-accept-screenshot`. A visual difference that the re-run reports is a finding, not something to sign off here.
* Do NOT call `ghost-inspector-execute-suite`. Run the one test, not everything around it.
**Validation:**
* If the test name cannot be determined from the summary, the comments or the {test-name} parameter, stop and say which of the three were tried.
* If `ghost-inspector-execute-test` returns no pass state, report the run as inconclusive and post no comment.
* If the issue is already Done, Closed or Resolved, say so and ask whether to proceed before running the test.Who it's for
One answer, three teams that use it
QA engineers
Verify fixes before closing
Rerun the test behind a bug and record the result on the issue, in one request.
Verify the fix for MOB-88 by rerunning its Ghost Inspector test.
Developers
Prove your fix works
Attach a fresh passing run to the bug before you hand it over for closure.
Rerun the Ghost Inspector test for WEB-1107 and post the result on the issue.
Release managers
Stop bugs coming back
Bugs carry run evidence before release, so regressions are caught before they ship.
Does the fix for API-219 hold? Run the Guest checkout test in Ghost Inspector and record it.
How it works
What PopdockAI does when you ask
You give the Jira issue key and your AI assistant finds and reruns the test behind the bug. The result goes on the issue as one comment, and the bug's status is left to you.
01
Ask
You give the Jira issue key.
02
Find
It reads the bug and finds its test, stopping if several match.
03
Run
It runs the test once and gathers the failing step if it fails.
04
Record
It posts one verification comment and never changes the status.
What PopdockAI provides
The tools behind the answer
Ready made tools for Jira and one secure connection to your AI tool. No code, no exports.
jira-get-issue
jira-get-issue-comment
ghost-inspector-get-test-result
ghost-inspector-get-test
ghost-inspector-execute-test
ghost-inspector-get-test-step
ghost-inspector-get-console-log
jira-add-issue-comment
jira-get-issue-watcher
This report is built from Get data tools only. PopdockAI applies the row, field and parameter rules your admin sets, and every query is written to the audit trail.
The outcome
Fixes proven with a fresh run
Related use cases
More ways to put this data to work
One prompt template with infinite possibilities.
Your AI assistant adapts its approach, tone, and capabilities to match each unique context while you maintain a single, manageable workflow.
Join forward-thinking companies that are transforming their operations with intelligent, secure AI automation.

.webp)
