📊
DataPython

Task Breakdown

by centminmod

Task Breakdown is a Data skill for Claude Code, published by centminmod in my-claude-code-setup.

2.6K stars246 forkson centminmod/my-claude-code-setupAdded 2026/08/18+1% in starsRepository updated 2026/08/01
claudeclaude-aiclaude-codeclaudecode-configclaudecode-hooksclaudecode-subagentssubagents
Install in seconds
Install Task Breakdown
Copy Task Breakdown into your Claude Code skills folder. Run the command in your terminal, or review the source on GitHub before installing.
terminal
npx degit https://github.com/centminmod/my-claude-code-setup/tree/master/.claude/skills/task-breakdown ~/.claude/skills/task-breakdown

Requires Node.js. Downloads this skill only — not the rest of the repository — into your Claude Code skills folder.

Without Node.js

git clone https://github.com/centminmod/my-claude-code-setup.git

Clones the whole repository, then copy the skill’s own directory into your skills folder yourself.

In this catalog

Source file
.claude/skills/task-breakdown/SKILL.md in centminmod/my-claude-code-setup
Installs to
~/.claude/skills/task-breakdown
Collection
One of 7 skills cataloged from this repository
Category
Data668 skills

What Task Breakdown does

Task Breakdown groups a session's turns into higher-level semantic tasks and renders a companion page with a worth-it, mixed, or likely-waste verdict per task. Use it after a session-metrics JSON export to see which work was worth it versus wasted.

Task Breakdown is cataloged under Data on DirSkills. Task Breakdown comes from a repository tagged claude, claude-ai, claude-code, claudecode-config and claudecode-hooks.

Documentation

README

Task Breakdown

Turns a session's per-request breakdown (the deterministic request_units emitted by session-metrics) into semantic tasks the user actually thinks in — "added auth", "debugged the cache miss" — and labels each with a verdict. You do the one thing deterministic code can't: decide which requests belong to the same task. The script does everything else (cost, turns, tokens, waste signals, the themed page).

This is the opening of the README. Read the full README on GitHub.

Commands Task Breakdown provides

Slash commands named in this skill’s SKILL.md, listed in the order they first appear.

  • /model
  • /debug
  • /feature-dev

Frequently asked about Task Breakdown

  • What else does centminmod publish alongside Task Breakdown?

    Task Breakdown is one of 7 skills that DirSkills catalogs from centminmod/my-claude-code-setup, the repository it ships in. Its siblings there include AI Image Creator, Audit Session Metrics and Claude Docs Consultant. Each one is a separate skill with its own page in this directory, installs the same way Task Breakdown does, and is maintained by centminmod in that same repository. The rest of the collection is listed on the centminmod/my-claude-code-setup page.

  • How does Task Breakdown compare to other Data skills?

    Task Breakdown ranks #317 by stars among the 668 Data skills in this catalog. The most-starred ones next to it are Benchmark Methodology, Jupyter Notebook and Solana. DirSkills ranks by the star count of the repository each skill ships in, so that order reflects how popular those repositories are rather than any review of Task Breakdown against them. Open each page to compare what they document and how they install.

More from centminmod/my-claude-code-setup

Task Breakdown is one of 7 skills cataloged on DirSkills from centminmod/my-claude-code-setup.

See all 7 skills
🎨
2w ago

AI Image Creator

AI Image Creator generates, edits, and analyzes images with AI models via OpenRouter and Cloudflare AI Gateway. Use it when a user asks to create a PNG, design an icon/logo/banner, describe an image, or analyze a video into a text prompt.
AI Engineering
2.6K246
📊
2w ago

Audit Session Metrics

Audit Session Metrics examines session-metrics JSON exports and produces a prioritized plain-English report of token-usage waste. Use it after an HTML export or when asked to audit, review, or find waste in saved session-metrics JSON.
Data
2.6K246
📚
2w ago

Claude Docs Consultant

Claude Docs Consultant fetches official Claude Code documentation on-demand from code.claude.com, using selective fetching for hooks, skills, subagents, plugins, MCP, CI/CD, IDE extensions, deployment, and more. Use it when a task requires authoritative guidance for any Claude Code feature.
AI Engineering
2.6K246
🤝
2w ago

Consult Codex

Consult Codex pairs OpenAI Codex with Claude's code search to analyze code questions from two perspectives with citation-verified evidence. Use it when you need a quick second opinion, debugging help, or architecture review with file:line references.
AI Engineering
2.6K246
🤖
2w ago

Consult Zai

Consult Zai pairs z.ai GLM 5.2 with Claude code-searcher to give a quick dual-model second opinion on code questions. Use it for complex code analysis, debugging, architecture questions, and code review where multiple perspectives help.
AI Engineering
2.6K246
📊
2w ago

Session Metrics

Session Metrics tallies Claude Code session token usage and cost estimates from the raw JSONL conversation log. Use it when asked about session cost, token usage, API spend, cache hit rate, or a breakdown of how much a Claude Code session has cost.
Data
2.6K246