/
/
See which test failures are new since the release
For operations and release teams who want Ghost Inspector failures lined up against Azure DevOps deployments, by asking their AI assistant.
On this page
Connector
Connect Ghost Inspector, pick your AI tool and paste the prompt.
The problem
Old failures get blamed on the new release
After a release, someone compares the Azure DevOps deployment time with Ghost Inspector runs by hand. It is easy to confuse failures that were already there with ones the release caused.
Without PopdockAI
- Find the deployment time in Azure DevOps
- Open Ghost Inspector runs before and after it
- Compare failing tests by hand
- Chase failures the release did not cause
With PopdockAI
- Release deployments and completion times are read
- Test failures before and after each deployment are compared
- New and existing failures are listed apart
- Results are correlation by time, and nothing is changed
The prompt
Use this prompt
Copy it into any AI tool connected to PopdockAI. Swap the values in braces for your own.
**Objective:** Find out whether the deployment of release `{release-name}` in Azure DevOps was followed by Ghost Inspector test failures within `{window-hours}` hours (default 24).
**Step 1: Establish the deployment (azure-devops)**
Call `azure-devops-get-deployment` with `release-name` set to `{release-name}`. Note each deployment's `environment-name`, `deployment-status`, `queued-on` and `completed-on`. If no deployment matches, stop and say so. If the release has several environments, list them and use the earliest completed deployment per environment.
**Step 2: Look at tests after the deployment (ghost-inspector)**
For each deployment with a `completed-on` value, call `ghost-inspector-get-suite-result` with `date-created.from` set to `completed-on`, `date-created.to` set to `completed-on` plus `{window-hours}` hours, and `passing` set to false. Then call `ghost-inspector-get-test-result` with the same window and `passing` false.
**Step 3: Compare with the period before**
Call `ghost-inspector-get-suite-result` for the same length of time ending at `completed-on` so you can tell new failures from long-standing ones. A test is a new failure only if it failed after the deployment and did not fail before.
**Step 4: Report**
A table per environment: deployment status, completed time, number of failing suite runs and test runs after, number before, and the names of new failures. End with one line: no signal, or likely regression.
**Rules & Guidelines:**
* Azure DevOps tools (prefix azure-devops-) come from the azure-devops server; Ghost Inspector tools (prefix ghost-inspector-) come from the ghost-inspector server. Both servers must be connected.
* The two systems share no key. The link is time: match on the deployment's `completed-on` against the test run's `date-created`. Ghost Inspector tests are not tied to a release, so say this is correlation, not proof.
* Compare times in UTC.
* Do not call any write tool: `ghost-inspector-execute-test`, `ghost-inspector-execute-suite`, `ghost-inspector-execute-on-demand-test`, `ghost-inspector-update-test`, `ghost-inspector-update-suite`, `ghost-inspector-add-test-result-comment`, `ghost-inspector-add-suite-result-comment`, `ghost-inspector-accept-screenshot`, or any create, duplicate, cancel or update tool. This prompt only reads.
**Validation:**
* If `deployment-status` is not a success, say so and still report, but flag that nothing may have been deployed.
* If Ghost Inspector returns no runs in the window, report "no test runs in the window" rather than "no failures".Who it's for
One answer, three teams that use it
Release managers
Know if the release broke anything
New failures since the deployment are listed apart, so you can decide on a rollback with facts.
Did the Fabrikam Portal release in Azure DevOps introduce any new Ghost Inspector failures?
QA engineers
Skip failures that were already there
Existing failures are listed separately, so you focus on what changed.
Which Ghost Inspector failures in the Checkout suite are new since the last Contoso Web deployment?
DevOps engineers
Check each environment's deployment
Failures are compared around every deployment of the release, stage by stage.
Compare Ghost Inspector failures before and after each deployment of the Northwind API release.
How it works
What PopdockAI does when you ask
You name the release and the suite, and your AI assistant separates failures that are new since the deployment from old ones. It stops if no deployment matches.
01
Ask
You name the release and the Ghost Inspector suite.
02
Read
It reads the release deployments and their completion times.
03
Compare
It reads the test failures before and after each deployment.
04
Report
You get new failures and existing ones, with the time-based caveat.
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.
azure-devops-get-deployment
ghost-inspector-get-suite-result
ghost-inspector-get-test-result
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
Investigate the failures the release introduced
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)
