agentsclimarketplace
Sign in

Shelves Plan & spec

Task breakdown

399 rows from 215 repositories

Turning one large thing into a list of small ones.

What Task breakdown skills agree on

226 skill files read, by 221 of the 230 authors on this shelf whose files we hold, 2026-09-06

Counted by distinct author, so one author publishing three of these counts once. Where a claim sits in fewer files than authors, that is said: a claim held by forty authors across three files is one file people copied, not forty people who agreed. Near-identical wordings are grouped and the other wordings are shown, so the grouping is yours to check.

What they tell the agent to do

  1. Write acceptance criteria for every task32 of 221 in 30 filesalso worded as Give every task acceptance criteria; Give every task explicit acceptance criteria
  2. Add checkpoints every two to three tasks18 of 221 in 16 filesalso worded as Add checkpoints after every two to three tasks; Add verification checkpoints every two to three tasks
  3. Schedule high-risk tasks early16 of 221 in 14 filesalso worded as Place high-risk tasks early; Place high-risk tasks early to fail fast
  4. Slice work vertically into complete feature paths16 of 221 in 14 filesalso worded as Slice vertically, one complete feature path per task; Slice vertically, one feature path per task
  5. Include verification steps in every task15 of 221 in 13 filesalso worded as Give every task a verification step; Give every task verification steps
  6. Map dependencies between components14 of 221 in 12 filesalso worded as Map the dependency graph between components; map dependencies between sub-problems
  7. Get human approval before implementing14 of 221 in 12 filesalso worded as Get human review and approval before implementing; Get human approval before starting implementation
  8. Enter read-only plan mode before writing code10 of 221also worded as Enter plan mode before writing any code; Enter read-only plan mode before writing any code
  9. Read the spec and codebase before planning9 of 221 in 7 filesalso worded as Read the relevant code and docs before planning; Read the spec and relevant code first
  10. Map the dependency graph before ordering tasks8 of 221also worded as Map dependencies between components before ordering tasks; Map the dependency graph before creating cards
  11. Break any task touching more than five files8 of 221also worded as Break down any task touching five or more files; Break tasks larger than five files into smaller tasks
  12. List each task's dependencies7 of 221also worded as Declare dependencies for every task; Record each task's dependencies
  13. Order tasks so dependencies are satisfied first6 of 221 in 4 filesalso worded as Order tasks so dependencies are satisfied; Order tasks so dependencies are satisfied, foundations first
  14. Order tasks bottom-up along dependencies6 of 221also worded as Order implementation bottom-up by dependency; Order tasks bottom-up, foundations first
  15. Map dependencies and identify the critical path5 of 221also worded as Identify task dependencies and the critical path; Map blocking dependencies and the critical path

What they tell it not to do

  1. Do not write code during planning18 of 221 in 16 filesalso worded as Writing code during planning; Write no code during planning
  2. Do not start implementation without a written task list15 of 221 in 13 filesalso worded as Do not start implementation without a written task list; Start implementation without a written task list
  3. Do not use horizontal slicing11 of 221also worded as Never slice horizontally by technical layer; Avoid horizontal slicing
  4. Never give a single-point estimate6 of 221also worded as Never give a single-number estimate; Never emit a single-number estimate
  5. Never create tasks without acceptance criteria6 of 221also worded as Writing tasks without acceptance criteria; Do not plan tasks without acceptance criteria
  6. Never create circular dependencies4 of 221also worded as Create dependency cycles; never create dependency cycles
  7. Do not start tasks estimated over eight hours4 of 221also worded as tasks touching more than four files; Do not start tasks estimated over eight hours
  8. Do not proceed without a requirement description3 of 221also worded as do not proceed without user approval; never proceed without an explicit Approve
  9. Never present a single number without a range3 of 221also worded as Never show a point estimate without a range; Never present a single-figure estimate without a range
  10. Don't start executing without user confirmation3 of 221also worded as Do not build without explicit user confirmation; Do not auto-file without user confirmation

What they expect to be installed

  1. AskUserQuestion18 of 221
  2. npm test14 of 221 in 12 files
  3. npm run build13 of 221 in 11 files
  4. git10 of 221
  5. Glob9 of 221
  6. grep6 of 221
  7. TaskCreate6 of 221
  8. jq6 of 221
  9. Linear6 of 221
  10. Read5 of 221

What they ask it to produce

  1. Ordered task list with acceptance criteria11 of 221 in 9 filesalso worded as Task list with acceptance criteria; A written task list with acceptance criteria
  2. Implementation plan document10 of 221 in 8 filesalso worded as plan.md implementation plan document; An implementation plan document
  3. Dependency graph9 of 221also worded as dependency-graph.md; Dependency graph with phases
  4. TASKS.md7 of 221also worded as design.md and tasks.md; tasks-summary.md
  5. Task checklist at tasks/todo.md6 of 221also worded as todo.md flat task checklist; tasks/todo.md
  6. Checkpoint verification checklists5 of 221 in 3 filesalso worded as Verification checkpoints
  7. Plan document5 of 221also worded as Plan document written to the issue; Task plan document
  8. Mermaid dependency graph4 of 221

When Task breakdown authors say to reach for one

The situations these authors wrote into their own files, counted out of the same 221 authors, with the skills that name each one

How Task breakdown skills are built

381 skill directories by 205 authors, read from their repositories’ own file trees 2026-08-05

The middle bundle among those shipping files is 2 files, 10.5 KB beside SKILL.md

Counted by distinct author, same as above, so one author publishing forty template copies counts once. SKILL.md itself is not counted as a file, so a single-file skill is one where that file is the whole skill.

The shape

  1. SKILL.md is the whole skill122 of 205 authors, 258 of 381 skills
  2. files ship beside it83 of 205 authors, 123 of 381 skills
  3. executable scripts ship inside21 of 205 authors, 23 of 381 skills

The folders they converge on

  1. references/50 of 205 authors, 58 of 381 skills
  2. scripts/17 of 205 authors, 19 of 381 skills
  3. evals/13 of 205 authors, 18 of 381 skills
  4. assets/8 of 205 authors, 9 of 381 skills
  5. agents/6 of 205 authors, 7 of 381 skills
  6. examples/5 of 205 authors, 7 of 381 skills

  • Recursive decomposition

    massimodeluisa/recursive-decomposition-skill/plugins/recursive-decomposition/skills/recursive-decomposition Skill

    44★ repo

    Claude Code skill for handling long-context tasks through recursive decomposition

  • Business decomposition order quality

    spikeHongg/china-stock-research-skills/skills/business-decomposition-order-quality Skill

    15★ repo

    Open-source agent skills for China stock and A-share research: Chinese-first outputs, citation-first analysis, and orchestrated workflows.

  • Decompose

    echology-io/decompose/io.github.echology-io/decompose MCP server

    10★ repo

    Decompose text into classified semantic units. Authority, risk, attention. No LLM.

  • Systems thinking workflow architect

    bybaochaule/systems-thinking-workflow-architect Skill

    4★

    Systems Thinking Workflow Architect helps an agent decompose complex problems and design reviewable process artifacts such as workflow maps, SOPs, decision trees, lead-generation processes, bottleneck analyses, and agent flow plans.

  • Activity duration estimation

    Ldxs001/workbuddy-skills/skills/activity-duration-estimation Skill

    3★ repo

    一个包含agent和skill的集合：RAG、技能审查、周报生成、项目管理、质量监控...及技术方案/内部设计文档

  • Progressive estimation

    Enreign/progressive-estimation Skill

    3★

    An AI skill for estimating AI-assisted and hybrid human+agent development work. Research-backed formulas. PERT statistics. Calibration feedback loops. Zero dependencies.

  • Prd decomposer

    Sixz-AI/prd-workflow/prd_decomposer Skill

    2★ repo

    Claude Code skills for AI-assisted PRD execution — decompose requirements into atomic REQs, enforce brainstorming before coding, track status per-REQ, and maintain git discipline from first spec to final commit.

  • Project planner skill

    428alexander9/claude-skills-marketplace/project-planner-skill Skill

    no license2★ repo

    🛠️ Transform Claude into a System Architect, generating detailed project documentation for AI-assisted software development with clear boundaries and traceable requirements.

  • Task decomposition

    sumitaich1998/jarvisvr/agent-backend/tests/fixtures/skills/orchestration/task-decomposition Skill

    2★ repo

    An AI agentic operating system for mixed reality on the Meta Quest 3 — your own J.A.R.V.I.S. Multi-agent orchestration, multimodal perception (sight/hearing/gaze), 42 holographic widgets, 20 LLM providers.

  • Clickup task breakdown

    JoaoEquer/Oficina/skills/clickup-task-breakdown Skill

    2★ repo

    A lean agent harness for AI-assisted development — NestJS + TypeScript + Prisma + PostgreSQL. Skills, rules and commands extracted from real projects.

  • Problem solving systems thinking

    bybaochaule/problem-solving-systems-thinking Skill

    1★

    This skill helps an agent decompose complex problems, map workflows, identify decision points, and design an agent's end-to-end flow from intake through validation and final response.

  • Learn from project

    xchuan-li/learn-from-project Skill

    0★

    A Claude skill that decomposes an AI-built or unfamiliar project into learnable knowledge points — so you can explain it, trust it, and own it.

  • Agent task compiler

    Arnie016/fable-mythos/skills/agent-task-compiler Skill

    0★ repo

    Reusable Claude Skills, evals, and repair loops for turning giant prompts into compounding AI workflows.

  • Planner skills

    goffity/planner-skills Skill

    0★

    Universal task planning and breakdown skill for AI coding agents. Auto-detects monorepo vs single-repo. Supports GitHub Issues and Jira.

  • Problem decomposition

    yacafx/skills-library/skills/problem-decomposition Skill

    0★ repo

    Portable, reusable Agent Skills library for coding workflows.

  • Adaptive estimation

    selamy-labs/agent-skills/skills/adaptive-estimation Skill

    0★ repo

    Reusable public SKILL.md workflows for AI agents

  • Staged decomposition

    Juliusolsson05/staged-decomposition Skill

    0★

    An Agent Skill that stops coding agents jumping straight to the finish line. Forces a written A -> B -> C -> D decomposition before any implementation code.

  • Taskforge explore

    hashirventhodi/taskforge-skills/taskforge-explore Skill

    0★ repo

    taskforge-skills — an Agent Skills framework for the durable-Task AI engineering workflow: skills reason, a deterministic engine is the only writer of task state

  • Genesis decomposer

    Xiaoher-C/agentbnb/genesis-template/templates/skills/genesis-decomposer Skill

    32★ repo

    Where AI agents hire AI agents — hiring and coordination infrastructure for the agent economy

  • Break down

    arezous/pm-pilot/.claude/skills/break-down Skill

    19★ repo

    AI copilot for product managers. Built for Claude Code and Cursor.

  • Decompose

    teklabsdigital/x2-method/skills/decompose Skill

    no license14★ repo

    AI builds faster than anyone, with no skin in the game and no memory of yesterday. How do you govern that? With X2, decisions stay in files people own, unbreakable rules fail the build, code is disposable, and every human turn is counted. Done is an audited report, never the agent's word.

  • Decompose

    ninthwave-io/ninthwave/skills/decompose Skill

    8★ repo

    Orchestrate parallel AI coding into reviewable PRs.

  • Issue authoring

    kurone-kito/idd-skill/skills/issue-authoring Skill

    4★ repo

    🤖 An automated setup kit for implementing loop engineering based on issue-driven development in your repository; since 2026.5

  • 02 paper strategy

    lianxhcn/PXa2026a/skills/core/02-paper-strategy Skill

    no license3★ repo

    Lecture

  • Cad construction

    Ed3Design/ed3design-engineering-bundles/cad-design/skills/cad-construction Skill

    2★ repo

    Claude Code skill bundles for hardware/maker engineering: 8 skills across 3 plugins — parametric CAD (Fusion 360 MCP-Bridge, OpenSCAD), FDM/3D-printing workflows (BOM validation, embedded UI docs), embedded systems (Victron Cerbo Modbus onboarding). Sibling to ed3design-skill-bundles.

  • Openspecpm

    aks-builds/openspecpm/skill/openspecpm Skill

    1★ repo

    Spec-driven, BDD-shaped project management for AI agents. OpenSpec proposals → BDD specs → tracked work in GitHub / Azure DevOps / Jira / Linear / GitLab.

  • Backlog grooming

    risadams/ink-and-agency/skills/business-product/backlog-grooming Skill

    no license1★ repo

    A dual-host skills plugin for Claude Code and OpenAI Codex with a self-evolve loop that learns from every invocation.

  • Spec

    Rtx09x/agentic-skills/skills/spec Skill

    0★ repo

    High-agency agent skills for executable specs and end-to-end shipping.

  • Planning and task breakdown

    addyosmani/agent-skills/skills/planning-and-task-breakdown Skill

    85,040★ repo

    Production-grade engineering skills for AI coding agents.

  • Planning and task breakdown

    sickn33/agentic-awesome-skills/plugins/agentic-awesome-skills-claude/skills/planning-and-task-breakdown Skill

    44,658★ repo

    AAS Core is the local, agent-first control plane for complete catalog discovery, agent-owned selection, stack validation, and planning, backed by 1,987+ agentic skills. Includes CLI, local MCP, catalog, plugins, and Workbench.

  • Progressive estimation

    sickn33/agentic-awesome-skills/plugins/agentic-awesome-skills-claude/skills/progressive-estimation Skill

    44,658★ repo

    AAS Core is the local, agent-first control plane for complete catalog discovery, agent-owned selection, stack validation, and planning, backed by 1,987+ agentic skills. Includes CLI, local MCP, catalog, plugins, and Workbench.

  • Planning and task breakdown

    sickn33/agentic-awesome-skills/plugins/agentic-awesome-skills/skills/planning-and-task-breakdown Skill

    44,658★ repo

    AAS Core is the local, agent-first control plane for complete catalog discovery, agent-owned selection, stack validation, and planning, backed by 1,987+ agentic skills. Includes CLI, local MCP, catalog, plugins, and Workbench.

  • Progressive estimation

    sickn33/agentic-awesome-skills/plugins/agentic-awesome-skills/skills/progressive-estimation Skill

    44,658★ repo

    AAS Core is the local, agent-first control plane for complete catalog discovery, agent-owned selection, stack validation, and planning, backed by 1,987+ agentic skills. Includes CLI, local MCP, catalog, plugins, and Workbench.

  • Planning and task breakdown

    sickn33/agentic-awesome-skills/skills/planning-and-task-breakdown Skill

    44,658★ repo

    AAS Core is the local, agent-first control plane for complete catalog discovery, agent-owned selection, stack validation, and planning, backed by 1,987+ agentic skills. Includes CLI, local MCP, catalog, plugins, and Workbench.

  • Progressive estimation

    sickn33/agentic-awesome-skills/skills/progressive-estimation Skill

    44,658★ repo

    AAS Core is the local, agent-first control plane for complete catalog discovery, agent-owned selection, stack validation, and planning, backed by 1,987+ agentic skills. Includes CLI, local MCP, catalog, plugins, and Workbench.

  • Recipe create task list

    googleworkspace/cli/skills/recipe-create-task-list Skill

    30,263★ repo

    Google Workspace CLI — one command-line tool for Drive, Gmail, Calendar, Sheets, Docs, Chat, Admin, and more. Dynamically built from Google Discovery Service. Includes AI agent skills.

  • 00 Full empirical analysis skill StatsPAI

    brycewang-stanford/Auto-Empirical-Research-Skills/skills/00-Full-empirical-analysis-skill_StatsPAI Skill

    no license3,315★ repo

    🔬 A curated collection of 23,000+ agent skills for empirical research across 8 social science disciplines. | 精选 23,000+ AI Agent 技能库，覆盖8大社会科学学科的实证研究。CoPaper.AI 20分钟完成一篇可复现的规范实证论文，并支持用户上传 Skills。-- Maintained by CoPaper.AI from Stanford REAP.

  • Bayesian estimation

    brycewang-stanford/Auto-Empirical-Research-Skills/skills/11-James-Traina-compound-science/skills/bayesian-estimation Skill

    no license3,315★ repo

    🔬 A curated collection of 23,000+ agent skills for empirical research across 8 social science disciplines. | 精选 23,000+ AI Agent 技能库，覆盖8大社会科学学科的实证研究。CoPaper.AI 20分钟完成一篇可复现的规范实证论文，并支持用户上传 Skills。-- Maintained by CoPaper.AI from Stanford REAP.

  • Event studies

    brycewang-stanford/Auto-Empirical-Research-Skills/skills/27-dariia-m-my_claude_skills/event-studies Skill

    no license3,315★ repo

    🔬 A curated collection of 23,000+ agent skills for empirical research across 8 social science disciplines. | 精选 23,000+ AI Agent 技能库，覆盖8大社会科学学科的实证研究。CoPaper.AI 20分钟完成一篇可复现的规范实证论文，并支持用户上传 Skills。-- Maintained by CoPaper.AI from Stanford REAP.

  • Iv estimation

    brycewang-stanford/Auto-Empirical-Research-Skills/skills/67-econfin-workflow-toolkit/iv-estimation Skill

    no license3,315★ repo

    🔬 A curated collection of 23,000+ agent skills for empirical research across 8 social science disciplines. | 精选 23,000+ AI Agent 技能库，覆盖8大社会科学学科的实证研究。CoPaper.AI 20分钟完成一篇可复现的规范实证论文，并支持用户上传 Skills。-- Maintained by CoPaper.AI from Stanford REAP.

  • Speckit tasks

    foryourhealth111-pixel/Vibe-Skills/bundled/skills/speckit-tasks Skill

    2,656★ repo

    VibeSkills is a general-purpose Skill that automatically routes local Skills and intelligently orchestrates harness workflows.

  • Hyperflow plan

    jeremylongshore/claude-code-plugins-plus-skills/plugins/ai-agency/hyperflow/templates/antigravity/skills/hyperflow-plan Skill

    2,628★ repo

    425 plugins, 2,810 skills, 200 agents for Claude Code. Open-source marketplace at tonsofskills.com with the ccpi CLI package manager.

  • Backlog grooming assistant

    jeremylongshore/claude-code-plugins-plus-skills/skills/20-enterprise-workflows/backlog-grooming-assistant Skill

    2,628★ repo

    425 plugins, 2,810 skills, 200 agents for Claude Code. Open-source marketplace at tonsofskills.com with the ccpi CLI package manager.

  • Task decomposition

    a5c-ai/babysitter/library/methodologies/ccpm/skills/task-decomposition Skill

    1,665★ repo

    Babysitter enforces obedience on agentic workforces and enables them to manage extremely complex tasks and workflows through deterministic, hallucination-free self-orchestration

  • Work decomposition

    a5c-ai/babysitter/library/methodologies/gastown/skills/work-decomposition Skill

    1,665★ repo

    Babysitter enforces obedience on agentic workforces and enables them to manage extremely complex tasks and workflows through deterministic, hallucination-free self-orchestration

  • Story decomposition

    a5c-ai/babysitter/library/methodologies/maestro/skills/story-decomposition Skill

    1,665★ repo

    Babysitter enforces obedience on agentic workforces and enables them to manage extremely complex tasks and workflows through deterministic, hallucination-free self-orchestration

  • Work unit decomposition

    a5c-ai/babysitter/library/methodologies/metaswarm/skills/work-unit-decomposition Skill

    1,665★ repo

    Babysitter enforces obedience on agentic workforces and enables them to manage extremely complex tasks and workflows through deterministic, hallucination-free self-orchestration

  • Task decomposition

    a5c-ai/babysitter/library/methodologies/spec-kit/skills/task-decomposition Skill

    1,665★ repo

    Babysitter enforces obedience on agentic workforces and enables them to manage extremely complex tasks and workflows through deterministic, hallucination-free self-orchestration

  • Wbs generator

    a5c-ai/babysitter/library/specializations/domains/business/project-management/skills/wbs-generator Skill

    1,665★ repo

    Babysitter enforces obedience on agentic workforces and enables them to manage extremely complex tasks and workflows through deterministic, hallucination-free self-orchestration

  • Task decomposition

    rohitg00/skillkit/packages/core/src/methodology/packs/planning/task-decomposition Skill

    1,458★ repo

    Supercharge AI coding agents with portable skills. Install, translate & share skills across Claude Code, Cursor, Codex, Copilot & 40 more

  • Detailed system decomposition hierarchy

    ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt3.5_8/detailed-system-decomposition-hierarchy Skill

    no license551★ repo

    AutoSkill: Experience-Driven Lifelong Learning via Skill Self-Evolution

  • Equipment spare parts cost estimation

    ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt3.5_8/equipment-spare-parts-cost-estimation Skill

    no license551★ repo

    AutoSkill: Experience-Driven Lifelong Learning via Skill Self-Evolution

  • Ldr matrix decomposition algorithm implementation

    ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt3.5_8/ldr-matrix-decomposition-algorithm-implementation Skill

    no license551★ repo

    AutoSkill: Experience-Driven Lifelong Learning via Skill Self-Evolution

  • Speculative historical estimation

    ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt3.5_8/speculative-historical-estimation Skill

    no license551★ repo

    AutoSkill: Experience-Driven Lifelong Learning via Skill Self-Evolution

  • Detailed system decomposition hierarchy

    ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt3.5_8_GLM4.7/detailed-system-decomposition-hierarchy Skill

    no license551★ repo

    AutoSkill: Experience-Driven Lifelong Learning via Skill Self-Evolution

  • Equipment spare parts cost estimation

    ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt3.5_8_GLM4.7/equipment-spare-parts-cost-estimation Skill

    no license551★ repo

    AutoSkill: Experience-Driven Lifelong Learning via Skill Self-Evolution

  • Matlab ldr matrix decomposition

    ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt3.5_8_GLM4.7/matlab-ldr-matrix-decomposition Skill

    no license551★ repo

    AutoSkill: Experience-Driven Lifelong Learning via Skill Self-Evolution

  • Polars mstl decomposition data preparation

    ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt4_8/polars-mstl-decomposition-data-preparation Skill

    no license551★ repo

    AutoSkill: Experience-Driven Lifelong Learning via Skill Self-Evolution

  • Semantic triplet extraction and decomposition

    ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt4_8/semantic-triplet-extraction-and-decomposition Skill

    no license551★ repo

    AutoSkill: Experience-Driven Lifelong Learning via Skill Self-Evolution

  • Dynamic gas estimation via json rpc for ethers js v6

    ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt4_8_GLM4.7/dynamic-gas-estimation-via-json-rpc-for-ethers-js-v6 Skill

    no license551★ repo

    AutoSkill: Experience-Driven Lifelong Learning via Skill Self-Evolution

All 399 in the catalog