Claude Code Plugins

Community-maintained marketplace

Feedback

ticket-management

@RoM4iK/tinker-public
0
0

Use when creating or updating tickets. Covers ticket writing best practices, what to include, and what to avoid.

Install Skill

1Download skill
2Enable skills in Claude

Open claude.ai/settings/capabilities and find the "Skills" section

3Upload to Claude

Click "Upload skill" and select the downloaded ZIP file

Note: Please verify skill by going through its instructions before using it.

SKILL.md

name ticket-management
description Use when creating or updating tickets. Covers ticket writing best practices, what to include, and what to avoid.

Ticket Management

TICKET CREATION AND UPDATE RULES

When creating or updating tickets (e.g., when planning backlog tickets):

❌ DO NOT add "Files to Create" or "Files to Modify" sections ❌ DO NOT specify exact file paths in tickets ✅ DO describe the desired outcome and what needs to work ✅ DO provide context, constraints, and technical notes ✅ DO let workers discover which files need changes

WHY: File lists constrain worker thinking. When workers see "Files to modify: agents.rb", they may only touch that file and miss related changes needed elsewhere (e.g., MCP bridge files, permissions, etc.).

Example - BAD:

## Files to Modify
- agents.rb

## Files to Create
- .claude/skills/memory/SKILL.md

Worker thinks: "Only edit these specific files."

Example - GOOD:

## What You're Building
A memory skill that agents can use to store and retrieve knowledge. The skill should:
- Be discoverable by Claude Code runtime
- Expose search_memory and store_memory tools
- Be assigned to all agent types

## Technical Notes
Skills live in `.claude/skills/<name>/SKILL.md` and are assigned via `skills: []` array in agents.rb.

Worker thinks: "What files do I need to touch to accomplish this?" → Explores codebase → Finds all affected files.

TICKET WRITING GUIDELINES

Good Ticket Structure

  1. Title: Clear, actionable description of the goal
  2. Context: Why this work is needed, what problem it solves
  3. What You're Building: Desired outcome and behavior
  4. Acceptance Criteria: How to verify it works
  5. Technical Notes: Architecture hints, patterns to follow, constraints
  6. Related Work: Dependencies, related tickets, context links

What to Include

  • Functional requirements - What should the feature do?
  • Behavioral specifications - How should it behave in edge cases?
  • Integration points - What systems/components does this touch?
  • Testing guidance - What scenarios should be tested?
  • Quality requirements - Performance, security, accessibility needs
  • Reference implementations - Similar patterns in the codebase

What to Avoid

  • Prescriptive file lists
  • Step-by-step implementation instructions
  • Overly detailed code snippets (unless showing a required pattern)
  • Assumptions about which files will change
  • Technical solutions when only requirements are known

MANTRA

"Describe the destination, not the path."