# Triage GitHub Issues With Clear Replies

Maintainers often need to understand issue reports before they answer. This skill guides Claude, Codex, or Claude Code through diagnosis, classification, decision support, and reply drafting.

## Install

```bash
npx skillstore add yunshu0909/issue-triage
```

## Metadata

- Status: approved
- Slug: yunshu0909-issue-triage
- Skillstore revision: r1
- Version status: missing
- Tree hash: 4ee9ec5f3d67c74761628eed0b9cd683f537a4230db5f166bac227c807770822
- Author: yunshu0909
- GitHub username: yunshu0909
- License: MIT
- Repository: https://github.com/yunshu0909/yunshu\_skillshub/tree/master/issue-triage
- Ref: 64ca8af0f54a325752f08bd54e52151061ea659a
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: allowed
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Risk factors: external\_commands
- Quality score: 78
- Quality tier: bronze
- Public page: https://skillstore.pages.dev/skills/yunshu0909-issue-triage
- Manifest: https://skillstore.pages.dev/api/skills/yunshu0909-issue-triage/manifest

## Capabilities

- Collects issue context from a user-provided GitHub issue or repository.
- Guides code investigation from entry points to downstream handlers.
- Classifies reports as bugs, architecture limits, feature requests, or usage questions.
- Compares fix scope, testability, user impact, and workaround options.
- Drafts professional GitHub replies for maintainer review before posting.

## Use Cases

- Handle bug reports: Trace reported behavior through the codebase, identify likely causes, and prepare a clear maintainer response.
- Review feature requests: Separate new capability requests from defects, estimate implementation cost, and explain product tradeoffs.
- Support repository users: Turn incomplete user reports into structured questions, workaround guidance, and polite follow-up replies.

## Prompt Templates

### Summarize an issue

```
Use this skill to read the GitHub issue I provide. Summarize the user environment, reproduction steps, expected behavior, actual behavior, and missing information.
```

### Find the root cause

```
Use this skill to inspect the relevant code path for this issue. Show the files, line references, call flow, and the likely root cause.
```

### Classify and decide

```
Use this skill to classify this issue as a bug, architecture limit, feature request, or usage question. Recommend whether to fix it, defer it, or suggest a workaround.
```

### Draft a maintainer response

```
Use this skill to draft a GitHub reply based on the diagnosis and decision. Explain the current design, user impact, workaround, and next steps.
```

## Limitations

- It does not guarantee access to private repositories or issue content.
- It depends on the assistant having suitable repository context.
- It does not replace maintainer judgment on roadmap or priority decisions.
- It should not publish comments without explicit user confirmation.

## Best Practices

- Read the full issue before drawing conclusions.
- Trace the complete code path before naming a root cause.
- Get user confirmation before publishing any GitHub reply.

## Anti Patterns

- Do not classify an issue from the title alone.
- Do not blame users for unsupported workflows.
- Do not promise fixes without checking scope and test conditions.

## Security Audit

- Audited at: 2026-07-08T10:43:02.676\+00:00
- Summary: All static findings are false positives caused by Markdown inline code or code fences, not executable Ruby or shell backticks. The skill is readable Markdown with no obfuscated payload, prompt injection, or suspicious hidden intent found.

## Stats

- Views: 159
- Downloads: 11
- Favorites: 0
- Popularity score: 0