/
/
Confirm a Ghost Inspector fix with one deliberate rerun
For developers who want to know a deployed fix worked, with the failure explained if it did not, by asking their AI assistant.
On this page
Connector
Connect Ghost Inspector, pick your AI tool and paste the prompt.
The problem
Rerunning until you get a lucky pass
After a fix is deployed, someone reruns the failing Ghost Inspector test, waits, and digs into the steps again if it fails. Reruns get repeated in hope of a lucky pass, and the original context is lost.
Without PopdockAI
- Rerun the test and wait
- Rerun again when it fails
- Open the steps to work out why
- Lose the context of the first failure
With PopdockAI
- Exactly one test must match, and what will run is shown
- A warning says the run acts on a live site
- The test runs once, with no retries, after you confirm
- A pass shows its run time, a failure its step and console errors
The prompt
Use this prompt
Copy it into any AI tool connected to PopdockAI. Swap the values in braces for your own.
**Objective:** Rerun the Ghost Inspector test named `{test-name}` and confirm whether a fix has taken effect, optionally overriding the starting url with `{start-url}` and the browser with `{browser}`.
**Step 1: Identify the test**
Call `ghost-inspector-get-test` with `name.contains` set to `{test-name}`. If more than one test matches, list the matches with their suite and id and stop; do not guess which one to run. Keep the `id`, and note its current `passing` state so you can say whether this rerun changes anything.
**Step 2: Show what is about to run and get agreement**
State the test name, its suite, the browser and region that will be used, and the starting url. Running a test performs real browser actions against a live site, which may create or change data there. Ask the user to confirm before continuing. Do not proceed without an explicit yes.
**Step 3: Run it, once**
Call `ghost-inspector-execute-test` with `test-id` from step 1, plus `start-url` and `browser` only if those parameters were supplied. Do not set `immediate`, because this prompt exists to report the outcome and `immediate` returns before there is one. Call this tool exactly once. If it fails to return a result, report that and stop; do not retry.
**Step 4: Report the outcome**
If `passing` is true, report that the test now passes, with the execution time, and compare that time against the test's previous state from step 1.
If `passing` is false, diagnose it before reporting. Call `ghost-inspector-get-test-result` with the `test-id` and `passing` false to find the run you just made, take its `id`, then call `ghost-inspector-get-test-step` with that `test-result-id` and `passing` false, and `ghost-inspector-get-console-log` with the same `test-result-id` and `error` true. Report the failing step, its command and target, and any related console errors.
**Rules and guidelines:**
* `ghost-inspector-execute-test` is the ONLY tool in this prompt that may change anything, and it may be called once, after the confirmation in step 2.
* Do NOT call `ghost-inspector-execute-suite`, `ghost-inspector-update-test`, `ghost-inspector-duplicate-test`, `ghost-inspector-execute-on-demand-test` or any create tool.
* Do not rerun a failing test hoping for a different answer. One run, then report.
**Validation:**
* If `{test-name}` matches nothing, say so and list some test names from `ghost-inspector-get-test`.
* If the user declines the confirmation, stop without running anything.
Who it's for
One answer, three teams that use it
Developers
Confirm your fix worked
One deliberate run tells you pass or fail, with the failing step if it did not work.
Rerun the Password reset test in Ghost Inspector now that my fix is deployed.
QA engineers
Verify on a specific environment
Run against a start URL or browser of your choice and get a clear result.
Rerun the Checkout guest test against https://staging.fabrikam.com in Chrome to check the fix.
Release managers
Close the loop before release
A single confirmed rerun gives a clear answer before the release goes ahead.
Run the Login smoke test once to confirm this morning's fix before we release.
How it works
What PopdockAI does when you ask
You name the test and optionally a URL or browser, and your AI assistant runs it once after you confirm. If it fails, the failure is diagnosed straight away.
01
Ask
You name the test, with an optional start URL or browser.
02
Confirm
It shows what will run, warns it acts on a live site, and waits for your yes.
03
Run
It runs the test once, with no retries.
04
Report
A pass comes with its run time, a failure with the failing step and console errors.
What PopdockAI provides
The tools behind the answer
Ready made tools for Ghost Inspector and one secure connection to your AI tool. No code, no exports.
ghost-inspector-get-test
ghost-inspector-execute-test
ghost-inspector-get-test-result
ghost-inspector-get-test-step
ghost-inspector-get-console-log
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
Know if the fix worked after one 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)
