create-github-issues
About
This skill automatically creates structured GitHub issues from code review findings or task breakdowns. It groups related findings into logical issues and applies labels using standard templates with summaries and acceptance criteria. Use it after code reviews to convert findings into actionable, trackable issues.
Quick Install
Claude Code
Recommendednpx skills add pjt222/agent-almanac -a claude-code/plugin add https://github.com/pjt222/agent-almanacgit clone https://github.com/pjt222/agent-almanac.git ~/.claude/skills/create-github-issuesCopy and paste this command in Claude Code to install this skill
Documentation
造 GitHub 議題
自評見或任分結構建 GitHub 議題。化見列(自 review-codebase、security-audit-codebase 或手析)為含標、驗準、交引之正議題。
用
- 碼評生須跟之見列後
- 謀議識須為議之工項後
- 轉 TODO 或備為可跟 GitHub 議
- 批建須一致式與標之關議
入
- 必:
findings— 項列,各至少含標與述。理亦含:嚴、影檔、建標 - 可:
group_by— 如何批見為議:severity、file、theme(默:theme)label_prefix— 自建標前綴(默:無)create_labels— 是否建缺標(默:true)dry_run— 不建僅預(默:false)
行
一:備標
保諸需標存於庫。
- 列存標:
gh label list --limit 100 - 識見所需標(自嚴、階、或顯標欄)
- 未映則映嚴至標:
critical、high-priority、medium-priority、low-priority - 映階/題至標:
security、architecture、code-quality、accessibility、testing、performance create_labels為 true→建缺標:gh label create "name" --color "hex" --description "desc"- 用一致色:紅為 critical/security、橙為 high、黃為 medium、藍為 architecture、綠為 testing
得: 見引之諸標存於庫。無重建標。
敗: gh CLI 未證→告用行 gh auth login。標建拒(權不足)→不建而記何標缺。
二:組見
批關見為議以避議蔓。
group_by為theme→按階或類組見(諸 security→1-2 議、諸 a11y→1 議)group_by為severity→按嚴組(諸 CRITICAL→1 議、諸 HIGH→1 議)group_by為file→按主影檔組- 組內按嚴序(CRITICAL 先)
- 組過 8 見→按子題分子組
- 各組為一 GitHub 議
得: 議組集、各含 1-8 關見。議總數可理(全碼評典型 5-15)。
敗: 見無組備→退為一見一議。小集(< 10)可、大集則生過多議。
三:組議
以標模築各議。
- Title:
[Severity] Theme: Brief description— 如[HIGH] Security: Eliminate innerHTML injection in panel.js - Body 構:
## Summary One-paragraph overview of what this issue addresses and why it matters. ## Findings 1. **[SEVERITY]** Finding description (`file.js:line`) — brief explanation 2. **[SEVERITY]** Finding description (`file.js:line`) — brief explanation ## Acceptance Criteria - [ ] Criterion derived from finding 1 - [ ] Criterion derived from finding 2 - [ ] All changes pass existing tests ## Context Generated from codebase review on YYYY-MM-DD. Related: #issue_numbers (if applicable) - 施標:嚴標+題標+任自定標
- 見引具檔→於體提之(非為指派)
得: 各議含明標、含嚴徽之編見、驗準勾、合標。
敗: 體過 GitHub 議寸限(65536 字符)→分議為部、交引之。
四:建議
以 gh CLI 建議並報果。
dry_run為 true→印各議標與體而不建,止- 各組議建之:
gh issue create --title "title" --body "$(cat <<'EOF' body content EOF )" --label "label1,label2" - 記各建議之 URL
- 諸議建後印結表:
#number | Title | Labels | Findings count - 議須序→加交引:編首議提「Blocked by #X」或「See also #Y」
得: 諸議建成。結表含議號與 URL 印。
敗: 單議建敗→誌誤續。終報敗。常敗:證逾、標缺(create_labels 為 false)、網超時。
驗
- 諸見現於至少一議
- 各議含至少一標
- 各議含驗準勾
- 無重議(察標與存開議)
- 議數合見數(大集非 1:1)
- 結表含諸議 URL 印
忌
- 議蔓:一見一議生 20+ 議難理。積極組——全評 5-10 議為理想
- 缺驗準:議無勾不可驗畢。各見當映至少一勾
- 標亂:過多標濾無用。限於嚴+題、非一見一標
- 陳引:自舊評建議→建前驗見仍適。碼或變
- 忘試行:大見集→必先以
dry_run: true預。編謀易於閉 15 誤議
參
review-codebase— 生此技消之見列review-pull-request— 生 PR 範見亦可化為議manage-backlog— 建後組議為 sprints 與優先create-pull-request— 建引並閉議之 PRcommit-changes— 承解議之修
GitHub Repository
Related Skills
content-collections
MetaThis skill provides a production-tested setup for Content Collections, a TypeScript-first tool that transforms Markdown/MDX files into type-safe data collections with Zod validation. Use it when building blogs, documentation sites, or content-heavy Vite + React applications to ensure type safety and automatic content validation. It covers everything from Vite plugin configuration and MDX compilation to deployment optimization and schema validation.
polymarket
MetaThis skill enables developers to build applications with the Polymarket prediction markets platform, including API integration for trading and market data. It also provides real-time data streaming via WebSocket to monitor live trades and market activity. Use it for implementing trading strategies or creating tools that process live market updates.
creating-opencode-plugins
MetaThis skill helps developers create OpenCode plugins that hook into 25+ event types like commands, files, and LSP operations. It provides the plugin structure, event API specifications, and implementation patterns for JavaScript/TypeScript modules. Use it when you need to intercept, monitor, or extend the OpenCode AI assistant's lifecycle with custom event-driven logic.
sglang
MetaSGLang is a high-performance LLM serving framework that specializes in fast, structured generation for JSON, regex, and agentic workflows using its RadixAttention prefix caching. It delivers significantly faster inference, especially for tasks with repeated prefixes, making it ideal for complex, structured outputs and multi-turn conversations. Choose SGLang over alternatives like vLLM when you need constrained decoding or are building applications with extensive prefix sharing.
