Connectors

/

GitHub

/

Comment failed GitHub builds on Jira issues
Write
Cross-Connector

Tell Jira issues when their GitHub build fails

For developers who want failed feature-branch builds noted on the right Jira issue, without pasting run details by hand.

Try PopdockAI free
Get a live demo
Comment failed GitHub builds on Jira issues
Post failed GitHub builds from the last 3 days in contoso/web to the matching issues in the WEB Jira project.
2 failed runs match open WEB issues; 1 matched issue is closed and 1 run has no key. - WEB-1042: CI failed on feature/WEB-1042-coupons, run 4820, Jordan Lee - WEB-1051: Build failed on WEB-1051-search, run 4817, Sam Patel - Skipped: WEB-1038 is closed - Unmatched: run 4811 on hotfix/footer Shall I post these two comments?
GitHub
Sample figures. Your results come from your own account.
github-get-workflow-run
jira-get-issue
jira-get-issue-watcher
jira-add-issue-comment
jira-get-issue-comment

On this page

The problem

The problem

The prompt

Who it's for

How it works

Tools used

Connector

GitHub

GitHub
Run it on your own GitHub data.

Connect GitHub, pick your AI tool and paste the prompt.

Try PopdockAI free

The problem

Jira never hears that the build failed

When a GitHub Actions build fails on a feature branch, the Jira issue for that work rarely hears about it. Linking them means finding the key in the branch or commit, opening Jira and pasting run details by hand.

Without PopdockAI

  • Spot the failed run in GitHub
  • Find the Jira key in the branch or commit
  • Open the issue and paste the run details
  • Post the same failure twice

With PopdockAI

  • The newest failed run per branch is found with its Jira key
  • Each key is confirmed, and closed issues are skipped
  • Issues already commented for that run are skipped
  • Comments post only after you confirm

The prompt

Use this prompt

Copy it into any AI tool connected to PopdockAI. Swap the values in braces for your own.

Prompt
**Objective:** Find the failed GitHub Actions runs in repository `{repository}` (if blank, use the connector's default repository) from the last `{lookback-days}` days (default 3), match each to a Jira issue in project `{project-key}`, and post the failure as a comment on that issue after the user confirms.

**Step 1: Read the failures**

Call `github-get-workflow-run` with `conclusion` = failure, `created-date.from` = today minus `{lookback-days}` days and `max-records` 50 (pass `company` = `{repository}` when given). If nothing comes back, report no failures and stop. Keep only the newest failed run per branch.

**Step 2: Find the Jira key**

For each run, look for a `{project-key}` key (uppercase, hyphen, number, for example ABC-123) in the branch first, then the commit-message. A run with no key goes in the unmatched list.

**Step 3: Resolve the Jira issue**

For each key, call `jira-get-issue` with `issue-key` set to it. If it does not exist, move the run to the unmatched list. Otherwise call `jira-get-issue-watcher` with `issue-key` set to it to get `issue-id`; if no row comes back, move it to the unmatched list with the reason no issue id.

Skip any issue whose status is Done or another closed status, and say so.

**Step 4: Confirm**

List each proposed comment (Jira key, workflow name, branch, run url) and the unmatched runs. Wait for an explicit yes.

**Step 5: Write**

For each confirmed issue, call `jira-add-issue-comment` once with `issue-id` and a body naming the workflow, the branch, the run-number, the short sha, the commit author and the run url.

**Step 6: Report**

Comments posted, and the unmatched runs with their reason.

**Rules & Guidelines:**

* `github-get-workflow-run` is on the github server; Tools whose names start with jira- are on the jira server.
* The only write is `jira-add-issue-comment`, at most one per Jira issue per run. Do NOT call `github-rerun-workflow-run`, `github-cancel-workflow-run` or `jira-update-issue`.

**Validation:**

* Before commenting, call `jira-get-issue-comment` with that `issue-id` and skip the issue if a comment already contains the run url, so a rerun of this prompt does not post twice.
* If a branch and its commit message name different keys, use the branch key and mention the other in the comment.

Who it's for

One answer, three teams that use it

Developers

Keep the ticket in sync with the build

Each comment names the workflow, branch, run, commit and author.

Post failed GitHub builds from today in fabrikam/api to the matching APP Jira issues.

Project managers

See which work has a broken build

Failed builds show up on the Jira issues you track, without asking developers.

Which WEB Jira issues had a failed GitHub build this week? Comment the details on them.

Tech leads

Avoid noise on closed work

Closed issues and runs already commented are skipped, so nothing posts twice.

Comment failed contoso/mobile builds from the last 5 days on the MOB Jira issues, skipping duplicates.

How it works

What PopdockAI does when you ask

You name the repository, the Jira project and a period, and your AI assistant drafts a comment for each matching issue. Nothing posts until you say yes.

01

Ask

You name the repository, the Jira project and a lookback period.

Your AI tool

02

Match

It finds the newest failed run per branch and reads a Jira key from the branch or commit.

github-get-workflow-run

03

Check

It confirms each issue, skipping closed ones and runs already commented.

jira-get-issue
jira-get-issue-comment

04

Confirm

You see the proposed comments, and they post only on your yes.

Your AI tool

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.

get-data

github-get-workflow-run

Returns GitHub Actions workflow runs with their status and conclusion. Filter by workflow name, run number, branch, triggering event, status, conclusion and created date range.
get-data

jira-get-issue

Returns Jira issues with their project, issue type, status, assignee, reporter, dates and estimates. Filter by project, status, assignee or date range to pull a backlog, a sprint or one person's open work.
get-data

jira-get-issue-watcher

Returns the users watching Jira issues, with the issue and project they are watching, so you can see who is following a piece of work.
run-action

jira-add-issue-comment

Adds a comment to a Jira issue, which is the usual way to record why something was reprioritised, reassigned or closed.
get-data

jira-get-issue-comment

Returns comments left on Jira issues with their author, body, created and updated dates. Filter by issue to read one conversation, or by date to see recent discussion across a project.
Read only.Nothing in GitHub Changes

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.

How Popdock AI secures your data

The outcome

Jira issues learn about their failed builds

Try PopdockAI free
Get a live demo

Related use cases

More ways to put this data to work

All Ramp use cases

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.

Try PopdockAI Free
See Documentation