数据生成
- 作者仓库星标 190,957
- 作者更新于 实时读取
- 作者仓库 n8n
- 领域
- 工程开发
- 兼容 Agent
-
- Claude Code
- Cursor
- Cline
- Codex
- Windsurf
- Gemini CLI
- +20
- 信任分
- 88 / 100 · 社区维护
- 作者 / 版本 / 许可
- @n8n-io · 未声明 license
- Token 消耗评级
- 低消耗
- 接入复杂程度
- 需简单配置
- 是否需要外部 API Key
- 不需要
- 兼容的系统
- macOS · Linux
- 底层运行要求
- Node.js
- 文件与系统权限
-
- 只读
- 允许写入 / 修改
- Shell 执行
- 网络行为
- 允许外网请求
- 安装命令数
- 26 条
档案由构建时根据 SKILL.md 与安装命令自动衍生,可能与作者实际意图存在差异。
需要注意: 未限定 allowed-tools,默认拥有全部工具权限。
---
name: n8n:create-issue
description: Create Linear tickets or GitHub issues following n8n conventions. Use when the user asks to crea…
category: 工程开发
runtime: Node.js
---
# n8n:create-issue 输出预览
## PART A: 任务判断
- 适用问题:代码实现、重构、调试或代码审查。
- 输入要求:目标材料、限制条件、期望输出和验收方式。
- 证据边界:围绕“Determine Target / Linear Tickets / Prerequisites”读取原文规则,不把推断写成作者承诺。
## PART B: 执行结果
- **01** 任务判断:确认你的需求是否属于代码实现、重构、调试或代码审查,并标出输入、限制和预期结果。
- **02** 执行计划:优先按“Determine Target / Linear Tickets / Prerequisites”拆成步骤,说明每一步会读取什么、修改什么、产出什么。
- **03** 交付结果:给出可复制的命令、文件改动、检查清单或内容草稿,并说明如何继续迭代。
- **04** 风险边界:结合 读取文件、写入/修改文件、执行终端命令、会按任务需要访问外部网络、通常不需要额外 API Key 给出执行前确认项。
## Running Rules
- 读取文件、写入/修改文件、执行终端命令;会按任务需要访问外部网络;通常不需要额外 API Key。
- 先小样例验证,再放大到真实任务。
- 交付时同时给结果、检查口径和下一步迭代建议。 原文没有稳定的斜杠命令要求。安装验证后通常全局生效,直接在对话里点名这个 Skill 并描述任务即可。
告诉 Agent 目标文件或材料、期望结果、不可改范围、是否允许联网或执行命令。本 Skill 的权限画像是:读取文件、写入/修改文件、执行终端命令。
先用一个小任务确认它会围绕“Determine Target / Linear Tickets / Prerequisites”工作;涉及文件或命令时,先看 diff、日志、预览或测试结果。
检查最终产物是否包含明确结果、必要证据和下一步动作;如果输出泛泛而谈,就补充输入、边界和验收标准后重跑。
---
name: n8n:create-issue
description: Create Linear tickets or GitHub issues following n8n conventions. Use when the user asks to crea…
category: 工程开发
source: n8n-io/n8n
---
# n8n:create-issue
## 什么时候使用
- 把工程方向的常用动作沉淀成 Agent 可调用的技能 适合处理工程开发场景下的代码实现、调试、重构、测试或代码审查,核心价值是把输入、判断、执行、验证和交付边界固定下来,避免 Agent 泛泛回答。 把任务拆成可执行、可检查、可继续迭代…
- 面向代码实现、重构、调试或代码审查,优先处理能明确输入、步骤和验收标准的工作。
## 需要提供什么
- 目标材料、目录范围、期望结果和不可改动内容。
- 是否允许联网、执行命令、读写文件或调用外部服务。
## 执行规则
- 围绕「Determine Target / Linear Tickets / Prerequisites」组织步骤,不把推断写成作者事实。
- 读取文件、写入/修改文件、执行终端命令;会按任务需要访问外部网络;通常不需要额外 API Key。
- 先跑小样例,确认结果可检查后再扩大任务范围。
## 输出要求
- 给出最终产物、关键证据、验证方式和下一步动作。
- 信息不足时标记 unknown,不编造命令、平台或依赖。 作者原文负责流程事实;仓库文件负责来源和命令;流狐只补充适用场景、限制和质量判断。
skill "n8n:create-issue" {
输入层 -> 用户目标 + 目标文件 + 禁止范围 + 验收标准
上下文层 -> Determine Target / Linear Tickets / Prerequisites
规则层 -> SKILL.md 触发条件 / 执行顺序 / 输出格式
运行层 -> Node.js | 读取文件、写入/修改文件、执行终端命令 | 会按任务需要访问外部网络
安全层 -> 通常不需要额外 API Key + 小任务验证 + diff / 日志复核
输出层 -> 可复制结果 + 检查清单 + 下一步迭代
} Create Issue
Create a Linear ticket or GitHub issue for: $ARGUMENTS
Determine Target
Decide where the issue should be created based on user input:
- If the user says "Linear", "ticket", or provides a team key (e.g., AI, NODE, N8N) → Linear
- If the user says "GitHub", "GH issue", or "open source" → GitHub
- If ambiguous, ask the user which platform they want
Linear Tickets
Prerequisites
Verify the Linear MCP is connected before proceeding.
Style Guide
Title
- Sentence case — capitalize only the first word (e.g., "Add webhook verification to Trello trigger")
- Descriptive — a reader should understand the scope without opening the ticket
- 5–15 words — long enough to be specific, short enough to scan
- Imperative mood for features/enhancements — "Add ...", "Support ...", "Improve ..."
- Bug titles — prefix with
Bug -followed by a description of the symptom (e.g., "Bug - Pin data not updating after workflow edit") - No ticket IDs in titles — the identifier (AI-1234) is assigned automatically
- No trailing punctuation
Description
Structure the description using markdown headers. Use the appropriate template:
For bugs:
## Description
[Clear explanation of the problem]
## Expected
[What should happen]
## Actual
[What happens instead]
## Attachments
[Screenshots, videos, or screen recordings that illustrate the problem]
## Steps to reproduce
1. [Step-by-step reproduction]
## Additional context
- n8n version: [version]
- Database: [SQLite/PostgreSQL]
- Hosting: [cloud/self-hosted]
For features / enhancements:
## Summary
[One-paragraph overview of what this adds or changes]
## Problem
[What limitation or gap exists today]
## Proposed solution
[How it should work — technical approach if known]
## Out of scope
[Explicitly note what this does NOT cover, if helpful]
For tech debt:
## Summary
[What technical improvement is needed]
## Current state
[What the code/system looks like today and why it's problematic]
## Proposed improvement
[What the improved state should look like]
## Motivation
[Why this matters — maintainability, performance, developer experience, etc.]
## Scope
[What is included / excluded from this work]
For spikes / investigations:
## Goal
[What question are we trying to answer]
## Context
[Why this investigation is needed now]
## Expected output
[What deliverable is expected — RFC, PoC, decision document, etc.]
Attachments (Screenshots / Videos)
If the user provides screenshots, videos, or screen recordings:
- URLs — embed directly in the description using markdown image syntax (
) - File paths — if the user provides a local file path, ask them to upload it to a hosting service (e.g., GitHub, Imgur) or use
mcp__linear-server__create_attachmentto attach it to the Linear ticket after creation - Pasted images in conversation — describe what the image shows in the ticket description and note that a screenshot was provided. You cannot upload binary data directly.
Always mention in the description when visual evidence was provided, even if it cannot be directly embedded.
Priority
| Value | Level | When to use |
|---|---|---|
| 4 | Low | Nice-to-have, no user impact |
| 3 | Normal | Default — standard planned work |
| 2 | High | Blocks other work or affects users significantly |
| 1 | Urgent | Production-breaking, security vulnerability, data loss |
| 0 | None | Not yet assessed |
Guardrails:
- Default to Normal (3) unless the user explicitly states otherwise
- Never set Urgent (1) unless the user explicitly says "urgent", "P0", "production down", or "security vulnerability"
- Never set None (0) — always make a priority assessment. If unsure, use Normal (3)
Status
Guardrails:
- Never create issues in Triage status — Triage is for externally-reported issues that enter through automated pipelines (GitHub sync, support escalation). Agent-created tickets have known context and should skip triage
- Default to Backlog — use this when the issue is acknowledged but not yet planned for a sprint
- Use Todo only when the user indicates the work is planned for the current cycle or should be picked up soon
- Never set In Progress, Review, or Done at creation time
Team
- Try to fetch up-to-date team areas of responsibility from Notion using
mcp__notion__notion-search(search for "areas of responsibility" or similar). Use the fetched data to determine the best team for the issue. - If Notion MCP is unavailable or the lookup fails, fall back to these common teams:
Engineering(N8N),AI,NODES,Identity & Access(IAM),Catalysts(CAT),Lifecycle & Governance(LIGO),Cloud Platform,Docs(DOC) - Always ask the user which team if not obvious from context or the Notion lookup
- If the issue is node-specific, it likely belongs to
NODES - If it involves AI/LangChain nodes, it likely belongs to
AI
Labels
Apply labels from these groups as appropriate:
Type (pick one):
bug— something is brokenfeature— net-new capabilityenhancement— improvement to existing functionalitytech debt— internal quality improvementspike— time-boxed investigationdoc— documentation-only change
Area (pick if applicable):
frontend,backend,performance,testing,infra,DX,Security-Team
Source (pick if applicable):
Internal— created by team membersGitHub— originated from a GitHub issueSentry— originated from error monitoringZammad— originated from support
Bucket (pick if applicable):
- Use the relevant feature-area bucket (e.g.,
Credentials,Canvas/Node,RBAC,LangChain nodes,Form Trigger, etc.)
Guardrails:
- Always apply a type label — every ticket needs at least a type
- Do not apply triage-state labels (
Triage: Pending,Triage: Complete, etc.) — these are managed by triage automation - Do not apply release labels (
n8n@1.36.0, etc.) — these are managed by release automation - Do not apply
docs-automationlabels — these are managed by docs automation
Estimates
Only set an estimate if the user provides one or explicitly asks for one. Use t-shirt sizes:
| Size | Value | Approximate effort |
|---|---|---|
| XS | 1 | ≤ 1 hour |
| S | 2 | ≤ 1 day |
| M | 3 | 2–3 days |
| L | 4 | 3–5 days |
| XL | 5 | ≥ 6 days |
Creating the Ticket
Gather required fields — if any are missing, ask the user:
- Title
- Team
- Description (draft one from the user's input using the templates above)
Present a preview before creating — show the user:
- Title
- Team
- Status
- Priority
- Labels
- Description (abbreviated if long)
Wait for user confirmation — do not create until the user approves
Create the ticket using
mcp__linear-server__save_issue:title: <title> team: <team name> description: <markdown description> priority: <priority number> state: <status name> labels: [<label names>]Report back with the issue identifier and URL
Things to Never Do (Linear)
- Never create issues in Triage status
- Never set Urgent priority without explicit user instruction
- Never apply triage-state, release, or docs-automation labels
- Never set assignee unless the user explicitly asks
- Never set a cycle or milestone unless the user explicitly asks
- Never create duplicate issues — if the user describes something that sounds like it may exist, search first with
mcp__linear-server__list_issues
GitHub Issues
Prerequisites
Verify gh CLI is authenticated: gh auth status
Important Context
The n8n GitHub issue tracker (n8n-io/n8n) is bug-only. Feature requests and questions are redirected to the community forum. Blank issues are disabled — the bug template must be used.
Style Guide
Title
- Sentence case — same as Linear
- Descriptive of the symptom — what is broken, not what you want
- No prefixes required — do not add "Bug:" or "Bug Report:" (the template handles categorization)
- No trailing punctuation
Body
GitHub issues must follow the bug report template structure:
### Bug Description
[Clear explanation of the bug]
### Steps to Reproduce
1. [Step 1]
2. [Step 2]
3. [Step 3]
### Expected Behavior
[What should happen]
### Debug Info
[If available — output from Help > About n8n > Copy debug information]
### Operating System
[e.g., macOS 14.2, Ubuntu 22.04]
### n8n Version
[e.g., 1.72.1]
### Node.js Version
[e.g., 20.11.0]
### Database
SQLite / PostgreSQL
### Execution Mode
main / queue
### Hosting
n8n cloud / self hosted
Guardrails:
- Always include reproduction steps — issues without them get closed as
closed:incomplete-template - Include debug info if available — this is critical for triage
- Never file feature requests as GitHub issues — redirect the user to the community forum or suggest creating a Linear ticket instead
Labels
Do not manually apply labels when creating GitHub issues. The triage automation handles labeling:
triage:pendingis auto-appliedstatus:in-linearis auto-applied when synced
Creating the Issue
Verify it's a bug — if the user describes a feature request, inform them that GitHub issues are bug-only and suggest alternatives (Linear ticket or community forum)
Draft the issue using the template above, filling in fields from the user's input
Present a preview before creating — show the user:
- Title
- Body (abbreviated if long)
- Repository (default:
n8n-io/n8n)
Wait for user confirmation
Create the issue using
gh:gh issue create --repo n8n-io/n8n --title "<title>" --body "$(cat <<'EOF' <body content> EOF )"Report back with the issue number and URL
Things to Never Do (GitHub)
- Never file feature requests as GitHub issues
- Never create issues without reproduction steps
- Never manually apply labels — let automation handle it
- Never create issues in repositories other than n8n-io/n8n unless the user explicitly specifies
Cross-Linking
When both a Linear ticket and GitHub issue exist for the same problem:
- Linear → GitHub: Add the GitHub issue URL as a link attachment on the Linear ticket
- GitHub → Linear: Add
https://linear.app/n8n/issue/<TICKET-ID>in the GitHub issue body
If the user creates one and mentions the other exists, offer to add the cross-link.
先判断是否适合
作者设计意图
作者的方法与取舍
边界和复核