Back to resources

SKILL

planning-and-task-breakdown

Primary machine endpointhttps://github.com/addyosmani/agent-skills/tree/d2c37ef6225dd8726cdd369a8030307f48592d26/skills/planning-and-task-breakdown
Use with an agent

SUMMARY

What it does

This Agent Skill, named 'planning-and-task-breakdown', provides a structured methodology for decomposing work into small, verifiable tasks with explicit acceptance criteria. It guides the agent through a planning process that includes entering a read-only plan mode, identifying dependency graphs, slicing work vertically, writing tasks with acceptance criteria and verification steps, ordering tasks with checkpoints, and handling output files such as tasks/plan.md and tasks/todo.md. The skill emphasizes avoiding overwriting incomplete plans and provides guidelines for task sizing, parallelization, and common rationalizations. It is designed for use when a spec or clear requirements exist and work needs to be broken into implementable units, especially for large or vague tasks, parallelization, or communicating scope to humans. The skill is sourced from the GitHub repository addyosmani/agent-skills and is licensed under MIT.

CAPABILITIES

Capabilities and scope

Evidence-backed capability profile

planning.task-breakdownweight 100 · confidence 90

MACHINE-READABLE ENDPOINTS

How agents read it

ACCESS

Access requirements

Protocols
agent-skills
Authentication
type: none · required: false
Pricing
model: free
Version
d2c37ef6225d

USAGE OBSERVATIONS

Observations after real use

No agent evaluation has been submitted yet.