tasks

Break down an approved implementation plan into specific, actionable tasks.

12 stars

bylionbenjamin

View on GitHub Installation ↓

Best use case

tasks is best used when you need a repeatable AI agent workflow instead of a one-off prompt.

Break down an approved implementation plan into specific, actionable tasks.

Teams using tasks should expect a more consistent output, faster repeated execution, less prompt rewriting.

When to use this skill

You want a reusable workflow that can be run more than once with consistent structure.

When not to use this skill

You only need a quick one-off answer and do not need a reusable workflow.
You cannot install or maintain the underlying files, dependencies, or repository context.

Installation

Claude Code / Cursor / Codex

$curl -o ~/.claude/skills/tasks/SKILL.md --create-dirs "https://raw.githubusercontent.com/lionbenjamin/agent-templates/main/skills/tasks/SKILL.md"

Manual Installation

Download SKILL.md from GitHub
Place it in .claude/skills/tasks/SKILL.md inside your project
Restart your AI agent — it will auto-discover the skill

How tasks Compares

Feature / Agent	tasks	Standard Approach
Platform Support	Not specified	Limited / Varies
Context Awareness	High	Baseline
Installation Complexity	Unknown	N/A

Frequently Asked Questions

What does this skill do?

Break down an approved implementation plan into specific, actionable tasks.

Where can I find the source code?

You can find the source code on GitHub using the link provided at the top of the page.

SKILL.md Source

# Tasks Skill

Break down plans into actionable implementation tasks.

## When to Activate

This skill is relevant when:
- Converting implementation plans to tasks
- Creating task breakdowns
- Organizing work into phases
- Preparing for implementation

## Core Principles

### Actionable
- Every task must be clear
- Implementable without ambiguity
- Concrete deliverables
- No vague descriptions

### Granular
- Small enough to complete in one session
- 1-2 hours maximum
- Single responsibility per task
- Easy to verify completion

### Sequential & TDD-Ordered
- Respect dependencies
- You can't fetch data before API exists
- Foundation before features
- Within each phase, test tasks come BEFORE implementation tasks
- Write the test (RED), then implement to pass (GREEN), then refactor

### Comprehensive
- Don't miss the "small stuff"
- Include icons, types, tests
- Update documentation
- Complete feature coverage

## Quick Checks

When creating task breakdowns, verify:
- [ ] All plan items converted to tasks
- [ ] Tasks are specific and actionable
- [ ] Granular enough (1-2 hours each)
- [ ] Dependencies respected
- [ ] Every implementation task has a preceding test task in same phase
- [ ] No separate "Verification & Tests" phase at the end — tests are interleaved
- [ ] Phases: Foundation → Logic (test→impl) → UI (test→impl) → Integration (test→impl) → Refactor
- [ ] Database tasks come first
- [ ] API tasks before UI tasks
- [ ] Parallelizable tasks marked with [P]
- [ ] Blocking tasks marked with [B]
- [ ] No missing "small stuff" (icons, types, tests)
- [ ] Completing all tasks equals completing feature
- [ ] Build and test tasks included