/
/
Give a new GitHub issue the right labels, owner and milestone
For developers and maintainers who want issues triaged from real labels and history, by asking their AI assistant.
On this page
Connector
Connect GitHub, pick your AI tool and paste the prompt.
The problem
Quick triage brings invented labels and wrong owners
Triaging a new GitHub issue means reading the thread, choosing labels that exist, finding who handled similar issues and picking a milestone. Done quickly between other work, invented labels and wrong assignees creep in.
Without PopdockAI
- Read the issue and its comments
- Type labels from memory
- Guess who should own it
- Pick a milestone that is already stale
With PopdockAI
- The issue, comments, labels in use and open milestones are read
- Owners are suggested from who closed similar issues
- Each proposal comes with a one line reason
- Only what you approve is applied, then read back
The prompt
Use this prompt
Copy it into any AI tool connected to PopdockAI. Swap the values in braces for your own.
**Objective:** Triage issue `{issue-number}` in `{repository}`: work out the labels, assignee and milestone it should have, and apply them only after the user confirms.
**Step 1: Read the issue**
Call `github-get-issue` with company `{repository}` and issue-number `{issue-number}`. Then call `github-get-issue-comment` with the same issue-number to read the discussion. Note what it already has: labels, assignee, milestone-title.
**Step 2: Gather the options**
* Labels: call `github-get-issue-label` with company `{repository}` and no issue-number to see the label-name values already in use in the repository. Only propose labels from that list, spelled exactly as returned.
* Assignee: call `github-get-issue` with labels.contains set to the proposed label and status `closed`, and look at who resolved similar issues. Propose one of those logins.
* Milestone: call `github-get-milestone` with state `open` and propose the one whose title or due-date fits the issue. Keep its milestone-number.
**Step 3: Propose and confirm**
Show the user the current values, the proposed values, and a one-line reason for each. Ask for confirmation. Apply only what the user approves.
**Step 4: Apply**
* For each approved label not already on the issue, call `github-add-issue-label` once.
* For an approved assignee not already assigned, call `github-add-issue-assignee` once.
* For an approved milestone, call `github-update-issue` with issue-number and milestone-number only.
Then call `github-get-issue` again and show the final state.
**Rules & Guidelines:**
* Never write before the user confirms. Never call `github-update-issue` with state, title or body in this prompt.
* Do NOT call `github-add-issue-comment`, `github-lock-issue` or `github-unlock-issue`.
* Never invent a label that is not in use in the repository; if none fits, say so and propose none.
**Validation:**
* If the issue is not found, say so and stop.
* If the issue is closed, say so and ask whether to continue before proposing anything.Who it's for
One answer, three teams that use it
Maintainers
Triage new issues fast
Get proposals with reasons and apply them in one reviewed step.
Triage issue 418 in fabrikam/app.
Team leads
Assign the right person
Assignees are suggested from who closed similar issues, not guessed.
Who should own issue 97 in contoso/api, and which milestone should it go on?
Developers
Use the labels that already exist
Only labels in use are proposed, so the label list stays clean.
Suggest labels for issue 133 in northwind/web from the ones we already use.
How it works
What PopdockAI does when you ask
You name the issue, and your AI assistant proposes labels, an owner and a milestone. It never invents a label or edits the title, body or state.
01
Ask
You name the issue.
02
Read
It reads the issue, its comments, the labels in use, similar closed issues and open milestones.
03
Confirm
You see each proposal with a reason and approve what you want.
04
Apply
It adds only what is missing and reads the final state back.
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.
github-get-issue
github-get-issue-comment
github-get-issue-label
github-get-milestone
github-add-issue-label
github-add-issue-assignee
github-update-issue
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
The right labels, owner and milestone in one step
Related use cases
More ways to put this data to work
Close out a GitHub milestone
Ask your AI to close out a GitHub milestone and it moves open issues to the next milestone and closes it, only after you confirm the plan.
Learn More
Check GitHub milestone readiness with Ghost Inspector
Ask your AI if a GitHub milestone is ready and it weighs its open issues against the latest Ghost Inspector results for a Ready, Ready with risks or Not ready verdict.
Learn More
Close GitHub issues for fixed Bugsnag errors
Ask your AI to close GitHub issues for fixed Bugsnag errors and it pairs them by exact error id, then comments and closes only the issues you confirm.
Learn More
GitHub milestone status report
Ask your AI for a GitHub milestone status report and it calculates progress, flags overdue and at-risk milestones and names the open issues holding each one up.
Learn More
Triage open GitHub issues
Ask your AI to triage open GitHub issues and it scores them on ownership, labels, activity and age, then reads the top ten threads to say what each needs.
Learn More
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.