AgentSkillsCN

ticket-planning-workflow

基于代码库背景,精心撰写 Linear 工单的通用模式。

SKILL.md
--- frontmatter
name: ticket-planning-workflow
description: Patterns for creating well-researched Linear tickets with codebase context.
user-invocable: false

Ticket Planning Workflow Skill

Transform Linear tickets into actionable implementation plans through systematic investigation and collaborative planning.

Purpose

Transform uncertainty into clarity by:

  • Gathering complete ticket context from Linear
  • Exploring relevant codebase sections
  • Identifying patterns and conventions
  • Asking targeted clarifying questions
  • Creating detailed, actionable implementation plans

Planning Methodology

Phase 1: Ticket Intelligence Gathering

Fetch and Parse Ticket

bash
# Get complete ticket information
mcp__linear__get_issue --id "$ticket_id"

Extract Structured Information

Parse from ticket:

  • Type: Feature/Bug/Chore (from labels)
  • Priority: Urgency level
  • Acceptance Criteria: From description checkboxes
  • Related Context: Parent tickets, dependencies, blockers
  • Discussion Points: Key decisions from comments

Identify Unclear Items

Flag for clarification:

  • Ambiguous requirements
  • Missing acceptance criteria
  • Architectural decisions needed
  • Scope boundaries unclear

Phase 2: Deep Codebase Exploration

Step 1: Broad Search

bash
# Find existing related code
rg "keyword" --type ts --files-with-matches

# Find similar implementations
rg "pattern-keyword" --type ts -A 5

Step 2: Understand Architecture

For each relevant file:

  • Document current implementation
  • Note patterns used
  • Identify extension points
  • Check for project-specific requirements

Step 3: Find Patterns to Follow

Extract from codebase:

  • Error handling: Common patterns
  • Entity pattern: Data structures
  • Route pattern: API organization
  • Service pattern: Business logic organization

Step 4: Check Project Guidelines

bash
# Review CLAUDE.md for requirements
cat CLAUDE.md | grep -A 10 "relevant-section"

Phase 3: Collaborative Clarification

Question Framework

Present 2-4 questions at a time with:

  • Context: What you found in codebase
  • Options: Concrete choices (A/B/C)
  • Recommendation: Your suggested approach
  • Trade-offs: Pros/cons of each

Format:

markdown
### Question: [Topic]

**Context**: [What codebase exploration revealed]

**Options**:
A) **[Option Name]** - [Description]
   - Pros: [Benefits]
   - Cons: [Drawbacks]

B) **[Option Name]** - [Description]
   - Pros: [Benefits]
   - Cons: [Drawbacks]

**Recommendation**: Option [X] because [reasoning]

**Your choice?** (A/B)

Phase 4: Implementation Plan Generation

High-Level Overview

markdown
# Implementation Plan: {ticket-id} - {title}

## Overview
[1-2 sentence summary]

## Approach
[3-5 bullet points on technical approach]

## Decisions
[List of decisions made during clarification]

## Complexity
[Simple/Moderate/Complex with justification]

Implementation Steps

For each step:

markdown
### Step N: [Title] (Est. complexity)
**What**: [Brief description]
**File(s)**: [Paths]

**Key points**:
- [Critical implementation detail]
- [Pattern to follow]

**Dependencies**: Step [X]
**Testing**: [How to verify]

Testing Strategy

Define:

  • Unit tests: Coverage goals, test categories
  • Integration tests: Key scenarios
  • Manual testing: Verification checklist

Risk Assessment

RiskImpactMitigation
[Risk]High/Med/Low[Strategy]

Success Criteria

Checklist covering:

  • Functional requirements
  • Technical requirements
  • Testing requirements
  • Documentation requirements

Phase 5: Git Strategy

markdown
## Git Strategy

**Branch**: {type}/{ticket-id}-{sanitized-title}
**Base**: main or staging

**Commits** (logical units):
1. `feat({ticket}): [description]`
2. `test({ticket}): [description]`
3. `docs({ticket}): [description]`

Integration

Used by agents:

  • linear-ticket-planner - Autonomous ticket planning
  • linear-ticket-creator - Ticket creation with exploration

Quality Checklist

Before completing plan:

  • Ticket requirements fully understood
  • Codebase exploration thorough
  • All clarifying questions asked
  • Implementation steps are actionable
  • Testing strategy is comprehensive
  • Risks identified and mitigated
  • Success criteria clear and measurable
  • Git strategy defined