linear-setup
tjcages/linear · Agent Skill
使用Bootstrap或擴充Linear專案的追蹤,具有證據支援的反填、里程碑、相依性和儲存庫紀律。
前往安裝 ↓總覽
| 發布者 | |
|---|---|
| 儲存庫 | tjcages/linear ↗ |
| 技能資料夾 | linear-setup |
| 星 | 10 |
| 附帶檔案 | 7 |
| 授權 | MIT |
| 最近更新 | 2026年7月29日 |
| 連結 | 技能原始碼 ↗ · 儲存庫 ↗ |
Mozgi 檢查
我們在技能檔案中發現的內容。詳細資訊需訂閱。
安裝
技能就是一個含 SKILL.md 的資料夾。選擇你的代理,在專案根目錄的終端機(macOS、Linux 或 Windows 的 Git Bash)執行指令。
把提示貼到 Claude、Codex 或其他助理:它會自行檢查並安裝技能。
Install the “linear-setup” skill from mozgi.io (https://mozgi.io/skills/tjcages-linear-setup/) for me. 1. Download the archive: https://mozgi.io/skills/tjcages-linear-setup/download If it fails, take the skill folder from the source: https://github.com/tjcages/linear/tree/main/skills/linear-setup 2. Read SKILL.md and the other files. Check the skill is safe: no hidden commands, no sending data elsewhere, no scripts doing something other than described. If anything looks suspicious, stop and tell me. 3. Put the “linear-setup” folder into your agent's skills folder: .claude/skills/ for Claude Code, .agents/skills/ for Codex, .cursor/skills/ for Cursor, .gemini/skills/ for Gemini CLI. Inside the project if we are in one, otherwise in the home folder (~/.claude/skills/ etc.). 4. Make sure <skills folder>/linear-setup/SKILL.md exists. 5. Briefly tell me what this skill does and offer to try it on my task. If you cannot do a step yourself, tell me what to click or type.
git clone --depth 1 --branch main https://github.com/tjcages/linear.git /tmp/linear mkdir -p .claude/skills cp -r /tmp/linear/skills/linear-setup .claude/skills/linear-setup
重新啟動 Claude Code — 技能就會出現在清單中,手動呼叫:/linear-setup。要套用到所有專案,請放在 ~/.claude/skills。
git clone --depth 1 --branch main https://github.com/tjcages/linear.git /tmp/linear mkdir -p .agents/skills cp -r /tmp/linear/skills/linear-setup .agents/skills/linear-setup
Codex 會從目前資料夾往上到儲存庫根目錄,在 .agents/skills 中尋找技能;用 /skills 或 $linear-setup 呼叫。所有專案:~/.agents/skills。
git clone --depth 1 --branch main https://github.com/tjcages/linear.git /tmp/linear mkdir -p .cursor/skills cp -r /tmp/linear/skills/linear-setup .cursor/skills/linear-setup
Cursor 啟動時會載入技能;在 Agent 聊天輸入「/」呼叫。Cursor 也會讀取 .claude/skills 與 .agents/skills。所有專案:~/.cursor/skills。
git clone --depth 1 --branch main https://github.com/tjcages/linear.git /tmp/linear mkdir -p .gemini/skills cp -r /tmp/linear/skills/linear-setup .gemini/skills/linear-setup
任務符合時,Gemini CLI 會提議啟用技能,由你確認。所有專案:~/.gemini/skills。
安裝前請先檢視 SKILL.md 與相關檔案。點擊檔案即可預覽。
SKILL.md 內容
作者原文 · 唯讀Linear setup
Chat rules: RESPONSE.md ↗. Auth: AUTH.md ↗. Full method: METHODOLOGY.md ↗.
Step 0 — auth
Probe Linear MCP. On failure → AUTH.md steps for this agent; stop. Never invent API calls.
Operating order
Follow in sequence. Early gates exist because getting them wrong corrupts live workflows.
1. Already tracked? (§1a)
Search Linear by repo name and ask the user. Never assume blank slate.
- Nothing → fresh bootstrap; continue.
- Exists → STOP. Enumerate issues/milestones/labels/docs/status updates → summarize → per-category yes/no before any write. Extend conservatively. Scan repo for unmirrored roadmap docs (§4.7).
2. North Star (§1b)
README-as-vision is usually enough — say the call out loud. Question 5 (why track?) picks the shape — see METHODOLOGY goal→shape table.
3–4. Type + structure (§2–§3)
Suggested types: Product / Tool / Drop / Marketing. Milestones 1:1 with real phases. Separate engineering / launch / rollout tracks. Dates only with a real signal. Prefer §17 if shape is unclear.
5. Backfill (§4)
CHANGELOG > roadmap > git log. Feature-level default; per-release for launched packages. Done only with evidence. Doc “done” ≠ proof — corroborate (§4.8).
Writing (tickets)
Titles ≤8 words. Descriptions ≤3 sentences or ≤5 bullets. Comments ≤3 lines. Screenshots over prose (§10).
6. Wire (§4, §10, §11)
Real blockedBy/blocks. Surface UI-only items early (§11). Add README as Linear project document.
7. Install discipline (§5)
Write the tracking protocol into always-loaded agent instructions (CLAUDE.md / AGENTS.md / .cursor/rules/) with real team/project/milestone names. Bootstrap is not done without this.
8. Honesty (§6, §9)
Status updates at milestone/health moments only. atRisk before dates slip.
9. Done check (§18, §26)
Rubric ≥16/20 with no load-bearing 0, or full §26 audit.
After setup
Point user to always-on (if missing) via finish linear-tracking install. Optional Cursor weekly health: AUTOMATION.md ↗. One next action.
常見問題
linear-setup 技能有什麼用途?
使用Bootstrap或擴充Linear專案的追蹤,具有證據支援的反填、里程碑、相依性和儲存庫紀律。
如何安裝 linear-setup?
打開本頁的「安裝」區塊並選擇你的代理 — Claude Code、Codex、Cursor 或 Gemini CLI:那裡有把技能資料夾複製到正確位置的現成指令。也可以複製「透過助理」分頁中的提示,讓助理自行檢查並安裝技能。
linear-setup 適用哪些 AI 模型?
技能是純 Markdown 指示(Agent Skills 格式),因此 linear-setup 適用任何讀取 SKILL.md 的代理:Claude Code、Codex、Cursor、Gemini CLI、GitHub Copilot、OpenCode 等。只有在任務符合技能描述時,模型才會載入指示。
linear-setup 是免費的嗎?
此技能由 tjcages 公開發布,採用 MIT 授權。技能本身免費 — 你只需向 AI 供應商支付實際使用的 token 費用。
安裝技能安全嗎?
技能是代理會遵循的指示,安裝前請閱讀 SKILL.md 及相關檔案。技能資料夾中的腳本只有在代理執行時才會運行 — 留意它的動作,不要授予多餘權限。
技能和 MCP 伺服器或外掛有何不同?
MCP 伺服器或外掛為模型提供新工具 — 會執行並回傳結果的程式碼。技能提供的是知識與流程:如何處理任務、要遵循哪些步驟、好的成果長什麼樣子。它不需要伺服器或 API 金鑰。