Agent Task Decomposition FAQs

Question 1

Can this skill identify project bottlenecks?

Accepted Answer

Yes, by mapping dependencies (depends_on and blocks), it identifies the critical path of the project, highlighting the chain of tasks that determines the overall timeline.

Question 2

Does it include testing in the implementation plan?

Accepted Answer

Absolutely. The skill includes specific fields for acceptance criteria and testing commands in the JSON structure, ensuring every task is verifiable.

Question 3

What files does the Agent Task Decomposition skill produce?

Accepted Answer

It produces two main files: TASKS.json, which serves as a machine-parseable source of truth for automation, and TASKS.md, which provides a human-readable view for developers.

Question 4

How does the skill handle task complexity?

Accepted Answer

It uses Fibonacci story points (1, 2, 3, 5, 8). Tasks exceeding 5 points trigger refinement, and those over 8 points must be broken down further to ensure manageable execution.

Question 5

How do I use this skill with other AI agents?

Accepted Answer

The generated 'prompt' field within each task is specifically designed to be passed to coding agents, providing them with context, requirements, and technical constraints.

Agent Task Decomposition

主要功能

使用场景

Agent Task Decomposition

主要功能

使用场景