How do I install the Task skill?

Run npx skillmds@latest add majiayu000/task-2 in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Task?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

majiayu000 (@majiayu000) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Task by majiayu000 — Agent Skill

Name: Task
Rating: 4.2 (5 reviews)
Author: majiayu000

Task

Create well-defined task specifications from plans or user prompts. Use when user wants to define a task, create a work item, break down a plan into actionable work, or write a spec for implementation. Triggers on "create task", "define task", "write spec", "break this down", "make this actionable", or when moving from planning to execution.

majiayu000 Updated Sep 9, 2026 567 repo stars

File contents

Task Creation

Create well-defined task specifications that provide clear goals while leaving implementation open to interpretation. Output to .work/tasks/backlog/<name>.md.

Task Creation Workflow

1. Understand the Scope

Determine task type and relevant sections:

Full-stack feature: All sections applicable
Backend only: Skip UX Guidelines, focus on System Behaviors and Data
Frontend only: Skip Data & Contracts internals, focus on UX
Documentation: Summary, Acceptance Criteria only
Bug fix: Summary, Edge Cases, Acceptance Criteria
Refactor: Summary, System Behaviors, Acceptance Criteria

2. Gather Context

If working from a plan:

Read the plan from .work/plans/
Extract relevant decisions and constraints
Link back to the plan

If working from user prompt:

Ask clarifying questions about scope
Identify affected files and systems
Understand success criteria

3. Write the Task Spec

Create .work/tasks/backlog/<descriptive-name>.md using the template below.

Omit sections not relevant to the task type. A backend API task doesn't need UX Guidelines. A docs update doesn't need System Behaviors.

Task Template

See references/spec-template.md for the full template.

Core sections:

Summary: One paragraph describing the task and why it exists
User Journey: Entry point, happy path, exit criteria
UX Guidelines: What user sees, copy tone, accessibility
System Behaviors: Invariants, state transitions, policies
Data & Contracts: Storage needs, API/events (high-level)
Usage/Cost Accounting: What usage to track and how
Edge Cases: Failure modes and user-visible outcomes
Acceptance Criteria: Testable outcomes as bullet list

Writing Guidelines

Be Specific About What, Open About How

Good: "User can filter todos by status (all, active, completed)" Bad: "Implement filtering using React state with useMemo optimization"

Define the outcome, not the implementation.

Reference Concrete Touchpoints

Include:

Files that will likely need changes
API routes or contracts involved
Database tables affected
UI components to modify

Keep It Scannable

Use bullet points over paragraphs
Bold key terms
Keep acceptance criteria atomic and testable

Task Lifecycle

After creating the task:

Task starts in .work/tasks/backlog/
When work begins, move to .work/tasks/in-progress/
When complete, move to .work/tasks/completed/

Update task status in the file header as work progresses.

---name: task-2description: Create well-defined task specifications from plans or user prompts. Use when user wants to define a task, create a work item, break down a plan into actionable work, or write a spec for implementation. Triggers on "create task", "define task", "write spec", "break this down", "make this actionable", or when moving from planning to execution.--- # Task Creation Create well-defined task specifications that provide clear goals while leaving implementation open to interpretation. Output to `.work/tasks/backlog/<name>.md`. ## Task Creation Workflow ### 1. Understand the Scope Determine task type and relevant sections: - **Full-stack feature**: All sections applicable - **Backend only**: Skip UX Guidelines, focus on System Behaviors and Data - **Frontend only**: Skip Data & Contracts internals, focus on UX - **Documentation**: Summary, Acceptance Criteria only - **Bug fix**: Summary, Edge Cases, Acceptance Criteria - **Refactor**: Summary, System Behaviors, Acceptance Criteria ### 2. Gather Context If working from a plan: - Read the plan from `.work/plans/` - Extract relevant decisions and constraints - Link back to the plan If working from user prompt: - Ask clarifying questions about scope - Identify affected files and systems - Understand success criteria ### 3. Write the Task Spec Create `.work/tasks/backlog/<descriptive-name>.md` using the template below. **Omit sections not relevant to the task type.** A backend API task doesn't need UX Guidelines. A docs update doesn't need System Behaviors. ## Task Template See [references/spec-template.md](references/spec-template.md) for the full template. Core sections: - **Summary**: One paragraph describing the task and why it exists - **User Journey**: Entry point, happy path, exit criteria - **UX Guidelines**: What user sees, copy tone, accessibility - **System Behaviors**: Invariants, state transitions, policies - **Data & Contracts**: Storage needs, API/events (high-level) - **Usage/Cost Accounting**: What usage to track and how - **Edge Cases**: Failure modes and user-visible outcomes - **Acceptance Criteria**: Testable outcomes as bullet list ## Writing Guidelines ### Be Specific About What, Open About How **Good**: "User can filter todos by status (all, active, completed)" **Bad**: "Implement filtering using React state with useMemo optimization" Define the outcome, not the implementation. ### Reference Concrete Touchpoints Include: - Files that will likely need changes - API routes or contracts involved - Database tables affected - UI components to modify ### Keep It Scannable - Use bullet points over paragraphs - Bold key terms - Keep acceptance criteria atomic and testable ## Task Lifecycle After creating the task: 1. Task starts in `.work/tasks/backlog/` 2. When work begins, move to `.work/tasks/in-progress/` 3. When complete, move to `.work/tasks/completed/` Update task status in the file header as work progresses.

Frequently asked questions