From kt
Break down a given task into subtasks with time estimates. Analyzes the codebase and task requirements to produce a structured plan with estimated effort for each step.
How this skill is triggered — by the user, by Claude, or both
Slash command
/kt:breakdownThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Given a task description (from the user or the current conversation context), perform the following:
Given a task description (from the user or the current conversation context), perform the following:
Understand the task: Read relevant source files, configs, and documentation to fully understand the scope of the task.
Identify subtasks: Decompose the task into concrete, actionable subtasks. Each subtask should be:
Estimate effort: For each subtask, assign a complexity size based on the scope of changes involved:
Output format: Present the breakdown as a tree structure that reflects the hierarchy and dependencies of the work:
## Task Breakdown: <task title>
## Specifications
- <key requirement or behavior 1>
- <key requirement or behavior 2>
- <constraint or edge case>
...
## Subtasks
- [ ] <high-level phase 1> [L]
- [ ] <subtask 1.1> [S]
- [ ] <subtask 1.2> [M]
- [ ] <sub-subtask 1.2.1> [XS]
- [ ] <high-level phase 2> [M]
- [ ] <subtask 2.1> [S]
- [ ] <subtask 2.2> [S]
- [ ] <high-level phase 3> [S]
## Subtask Details
### 1. <high-level phase 1>
<what this phase accomplishes and why>
#### 1.1 <subtask 1.1>
<what to do, which files/modules are involved, acceptance criteria>
#### 1.2 <subtask 1.2>
...
### 2. <high-level phase 2>
...
Save option: After presenting the breakdown, ask the user whether they want to save the plan as plan.md in the current working directory. If the user agrees, write the breakdown to plan.md.
npx claudepluginhub kt3k/skills --plugin ktProvides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.