Copy-paste the block below into any AI coding agent to turn a backlog of untriaged issues into labeled, prioritized, answerable queues.
Keywords: issue triage, maintainer, labels, duplicates, prioritization, backlog
Triage the open issues in this repository. The goal is a backlog a maintainer
can act on in priority order, not a pile of "looked at it" noise.
## Steps
1. **Check reproducibility first** - Confirm clear repro steps, expected vs
actual behavior, and environment details. If one is missing, reply naming
that exact piece (not "please provide more info") and label
`needs-repro`/`needs-info` rather than guessing.
2. **Detect and link duplicates** - Search open and closed issues for the same
root cause before triaging as new. Link duplicates to the canonical issue,
summarize why they match, and close with a pointer rather than leaving both
open.
3. **Apply one label rubric** - Use the repo's actual label set; if none
exists, propose one (`severity: blocker/high/medium/low`) instead of ad hoc
labels, and justify each severity in one line. Label small, well-scoped
issues needing no deep repo context as `good first issue` and comment with
the file to start from.
4. **Apply the stale policy** - Follow the repo's stale-bot policy; if none
exists, propose one (ping after 60 days of no response, close 14 days later)
and apply it consistently, with a comment saying why and how to reopen.
5. **Summarize the pass** - Report counts by label and severity, the duplicates
merged, and the issues needing a maintainer decision you cannot make
(design, breaking changes, roadmap).
## Verification
- [ ] Reproducibility was confirmed for each bug report before it was labeled.
- [ ] Every duplicate links the canonical issue and the reasoning for closing.
- [ ] The repo's own label taxonomy and severity rubric were used, not an
invented one.
- [ ] Every good-first-issue candidate carries mentoring notes on how to
approach it.
- [ ] Stale issues follow the repo's stated policy and are labelled, not
silently closed.
- [ ] Severity that genuinely depends on product judgment was raised as a
question rather than assumed.
- [ ] The pass is summarized with counts by label and severity and a list
needing a human decision.
## Rules
- Never close an issue as invalid or duplicate without linking the reasoning or
the canonical issue - a silent close reads as dismissive and loses context
for whoever revisits it.
- Don't invent a label taxonomy that ignores one the repo already has; extend
it, don't fragment it.
- Ask, don't assume, when severity or priority genuinely depends on product
judgment rather than technical facts.