bew/dotfiles/write-github-issue
Claude Code and agent skill by bew/dotfiles
- community
Checking that the SKILL.md file is still in the repository.
Guidelines for drafting a GitHub issue in bew's voice. Legacy issue-only drafter, superseded by `draft-github-issue-pr`; kept for comparison. Not auto-triggered.
Install
npx vskill@latest install bew/dotfiles/write-github-issueSKILL.md
Skill: write-github-issue
Goal
Draft a GitHub issue that is clear, well-researched, and framed to lower maintainer friction.
Issue types
<!-- §issue-types -->
Supported types — only one scenario is currently specified:
- Ask for something (feature request, capability ask, ask-for-help) — see <./refs/ask-for-something.md>
Other types (RFC, bug report, question, discussion) — no guidelines yet. Apply general principles below and use judgment.
General principles
IMPORTANT: Load bew-communication-style before writing any prose — required for voice and tone.
- Show your work. Mention what you searched, tried, or considered before writing the issue.
Maintainers respond better when they see the effort.
- Be concrete. Name real APIs, config keys, methods, or behaviors.
Avoid abstract descriptions when specific ones are possible.
- Explain context before asking. Give the maintainer enough background to understand why
this matters before they reach the ask.
- Offer alternatives. Multiple implementation paths signal you've thought it through and
reduce the cognitive load of having to invent one.
- Keep scope tight. One issue = one concern.
Split if there are multiple independent asks.
- End with an optional broader angle only if genuinely relevant
(e.g. "is this something that belongs upstream instead?"). Don't force it.
- Be nice. Open source is a shared space — a genuine thank-you and a light touch of warmth go a long way.
1–3 smileys max across the whole issue; never forced.
Steps
- Identify the issue type — read the matching ref if one exists (see <§issue-types>).
- Context check — assess whether there is enough data to apply general principles:
- Is the problem grounded in real, tested behavior?
- Are alternatives explored (not just invented on the spot)?
If data is thin, ask the user: > Not enough context to write a well-grounded issue. > Should we stop and gather more first, or draft a lighter version for your own research notes?
Proceed only once user decides.
- Read the matching ref file (if one exists for this issue type), then draft the issue body
applying general principles and any type-specific guidelines.
- Review: every sentence must earn its place. Cut filler.
Security
- Security scanPassed, 95 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingopencode/skills/write-github-issue/SKILL.md
Popularity
- 7-day trend score
- 5
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/bew/dotfiles/write-github-issue)