/
/
Track a failed Ghost Inspector suite run in GitHub, once
For developers who want suite failures in GitHub with their evidence and a link back, without duplicate issues, by asking their AI assistant.
On this page
Connector
Connect GitHub, pick your AI tool and paste the prompt.
The problem
Duplicate issues for the same broken suite
Turning a failed Ghost Inspector suite run into a GitHub issue means copying failing steps and console errors, checking for an open issue and linking the two. Skip the check and the backlog fills with duplicates.
Without PopdockAI
- Open each failing test in the suite run
- Copy failing steps and console errors
- Search GitHub for an open issue
- Link the run and issue by hand
With PopdockAI
- Each failing test's step and first console errors are read
- An open issue tracking the suite is looked for
- A new issue or a comment is drafted for you to confirm
- Each system is written at most once, with a link on the suite run
The prompt
Use this prompt
Copy it into any AI tool connected to PopdockAI. Swap the values in braces for your own.
**Objective:** Record the failed Ghost Inspector suite result `{suite-result-id}` as a GitHub issue, applying the label `{label}` if given, and link the issue back on the suite result.
**Step 1: Read the failure (Ghost Inspector)**
Call `ghost-inspector-get-suite-result` with `id` set to `{suite-result-id}`. If it is not found, report that and stop. If `passing` is true, say the run passed and stop.
Call `ghost-inspector-get-test-result` with `suite-id` set to the result's `suite-id`, `passing` set to false, and `date-created.from` / `date-created.to` set to the suite result's `date-execution-started` and `date-execution-finished`. For each failing test result, call `ghost-inspector-get-test-step` with `test-result-id` and `passing` false to get the failing step, and `ghost-inspector-get-console-log` with `test-result-id` and `error` true for console errors (keep the first three per test).
**Step 2: Look for an existing issue (GitHub)**
The issue title is `[Ghost Inspector] <suite-name> failing`. Call `github-get-issue` with `title.contains` set to `[Ghost Inspector] <suite-name>` and `status` open.
**Step 3: Draft, confirm, then write (GitHub)**
Draft the body in markdown: the suite name, the suite result id `{suite-result-id}`, the run time, pass/fail counts, then one section per failing test with its name, test id, test result id, the failing step's command, target and value, and the console errors.
Show the draft to the user and say whether you will create a new issue or comment on the existing one. Wait for confirmation.
* If an open issue exists: call `github-add-issue-comment` with its `issue-number` and the draft as `body`.
* Otherwise: call `github-add-issue` with the title, the draft as `body`, and `{label}` as `label` if given. It returns nothing, so call `github-get-issue` with `title.contains` again to get the new `issue-number` and `issue-url`.
**Step 4: Link back (Ghost Inspector)**
Call `ghost-inspector-add-suite-result-comment` with `suite-result-id` set to `{suite-result-id}` and a `message` of `Tracked in GitHub issue #<issue-number>: <issue-url>`.
**Step 5: Report**
The issue number and url, whether it was new or commented on, and the failing tests included.
**Rules & Guidelines:**
* The ghost-inspector tools are on the ghost-inspector server; the github tools are on the github server.
* Write at most once to each system: one `github-add-issue` or one `github-add-issue-comment`, then one `ghost-inspector-add-suite-result-comment`, and only after the user confirms the draft.
* Do NOT call `ghost-inspector-execute-suite`, `ghost-inspector-cancel-suite-run`, `github-update-issue` or `github-add-issue-assignee`.
**Validation:**
* If the suite result has failing counts but no failing test results in its time window, still file the issue with the suite-level counts and say the per-test detail could not be matched.
* If the new issue cannot be found again after creating it, do not create a second one. Report that the issue was created but could not be linked, and skip Step 4.Who it's for
One answer, three teams that use it
QA engineers
File suite failures with evidence
Failing steps and console errors go into the issue, so developers can start straight away.
File a GitHub issue for today's failed Login suite run in Ghost Inspector.
Developers
Add to the existing issue
If an issue already tracks the suite, a comment is added instead of a new issue.
Add the latest failed Search suite run to the existing GitHub issue if there is one, labelled e2e.
QA leads
Link test runs and issues both ways
The suite result gets a comment with the GitHub link, so anyone can follow it.
Track last night's failed Nightly regression run in GitHub and link it back in Ghost Inspector.
How it works
What PopdockAI does when you ask
You give the suite result and an optional label, and your AI assistant drafts a new issue or a comment. Nothing is written until you confirm.
01
Ask
You give the suite result and an optional label.
02
Read
It reads each failing test's step and first console errors.
03
Check
It looks for an open GitHub issue already tracking the suite and drafts an issue or comment.
04
Confirm
On your yes it writes once and comments the GitHub link on the suite result.
What PopdockAI provides
The tools behind the answer
Ready made tools for GitHub and one secure connection to your AI tool. No code, no exports.
ghost-inspector-get-suite-result
ghost-inspector-get-test-result
ghost-inspector-get-test-step
ghost-inspector-get-console-log
github-get-issue
github-add-issue-comment
github-add-issue
ghost-inspector-add-suite-result-comment
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
Every suite failure tracked once
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)
