Why is this better than just writing code directly?

By generating a specification first, you ensure that the AI understands the full scope and technical constraints of a task, leading to fewer errors and more consistent implementations during the build phase.

What is the FABER Architect skill?

The FABER Architect skill is an AI-powered tool designed to convert project requirements into actionable technical specifications and design documents during the second phase of a FABER workflow.

How does it interact with Git?

The skill automatically commits the generated specification file to your current branch and can be configured to push that commit to your remote repository automatically.

Where does this skill save the generated specifications?

By default, specifications are saved as Markdown files in the .faber/specs/ directory within your project repository.

Does it support different project domains?

Yes, while it primarily defaults to engineering tasks, it includes configuration support for design, writing, and data domains to provide relevant architectural guidance.

FABER Architect

Name: FABER Architect
Author: fractary

$fractary$ byfractary

•

生产力与工作流

Generates comprehensive technical specifications and design documents from project requirements to streamline software implementation.

The FABER Architect skill serves as the blueprint generator for Claude Code workflows, bridging the gap between high-level requirements and concrete code implementation. It automatically analyzes work items—whether they are bugs, features, or chores—to produce detailed design documents, technical approaches, and clear success criteria. By ensuring every task is backed by a version-controlled specification within your repository, it eliminates ambiguity for developers and provides a transparent audit trail for all architectural decisions.

主要功能

01Support for domain-specific workflows (Engineering, Design, Writing, Data)

02Standardized FABER response formatting for automated workflow management

03Seamless Git integration for versioning design documents

04Definition of clear success criteria and acceptance testing standards

05Automated technical specification generation with requirements and approach

062 GitHub stars

使用场景

01Establishing success criteria for complex bug fixes to prevent regressions

02Designing new feature implementations with step-by-step technical blueprints

03Documenting architectural decisions and refactoring plans for legacy code

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add fractary/claude-plugins architect

For use in Claude.ai and ChatGPT

Download Skill