classify-complexity

Classify ticket complexity against the codebase using a four-level rubric.

1|1|Updated Feb 25, 2026
One-click install
npx skills add https://github.com/williamthorsen/codeassembly --skill classify-complexity-williamthorsen
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: classify-complexity
Source: https://github.com/williamthorsen/codeassembly/tree/main/packages/agents/content/skills/classify-complexity
Command: npx skills add https://github.com/williamthorsen/codeassembly --skill classify-complexity-williamthorsen

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Estimating how hard a ticket will be is often guesswork. This Skill grounds the estimate in evidence by examining the actual codebase the ticket touches, producing a structured complexity verdict with scope, drivers, and risks. ## Core Features & Use Cases - Codebase-grounded assessment: Verifies the files, modules, and APIs mentioned in a ticket against what actually exists in the repository. - Four-level rubric: Maps findings to trivial, mechanical, involved, or architectural levels, preferring the higher level when characteristics span two. - Structured output: Produces a Markdown report with a verdict emoji, scope breakdown, classification drivers, and identified risks, stamped with the current git SHA. - Use Case: Before sprint planning, run the Skill on issue #99 to learn it crosses three packages and modifies shared interfaces, classifying it as architectural so the team allocates appropriate review and design time. ## Quick Start Ask the assistant to classify the complexity of issue #99 against the current codebase.

Frequently Asked Questions about classify-complexity

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I classify the complexity of a GitHub issue?

Provide the issue URL or a shorthand reference like #99, and the Skill resolves the ticket, examines the affected code, and maps findings to a four-level rubric from trivial to architectural. The output includes scope, drivers, and risks.

What complexity levels does the ticket classification rubric use?

The rubric has four levels: trivial, mechanical, involved, and architectural. When a ticket's characteristics span two levels, the Skill classifies it at the higher level.

Can I classify a ticket without providing an issue URL?

Yes. The ticket source is optional and can be auto-resolved from the environment, including a URL previously stored in the branch manifest. You can also pass a file path or plain text description of the work.

Does the complexity classifier modify tickets or post comments?

No. The Skill only produces an assessment as output. It does not modify tickets, save artifacts, or post comments to any platform.

What evidence does a complexity assessment cite?

Every classification cites specific codebase evidence such as file paths, module counts, affected packages, and dependency references. The report is also stamped with the current git short SHA for traceability.