How do I move from one project phase to the next?

After completing all sessions in a phase, you use the Stage 3 commands like /audit, /pipeline, and /infra to finalize the phase before running /phasebuild to start the next one.

Does this skill support monorepo architectures?

Yes, it includes native monorepo support, allowing you to manage multiple packages from a single root .spec_system directory with package-specific metadata.

What is the core philosophy of Apex Spec Workflow?

It operates on the '1 session = 1 spec' principle, limiting work to 2-4 hours and 12-25 tasks to fit perfectly within AI context windows and human focus spans.

Can I customize the workflow scripts?

Yes, you can choose to copy scripts locally to .spec_system/scripts/ during initialization, which allows you to customize the automation logic for your specific environment.

Why does the system require ASCII-only characters?

Strict ASCII encoding prevents character encoding bugs and tokenization errors in AI models, ensuring that specifications and tasks are interpreted with 100% accuracy.

Apex Spec Workflow

Name: Apex Spec Workflow
Author: moshehbenavraham

bymoshehbenavraham

•

Productivity & Workflow

Manages large-scale software development through a structured, specification-driven workflow that breaks complex projects into manageable 2-4 hour sessions.

The Apex Spec Workflow is a comprehensive system designed to scale AI-assisted development by breaking complex projects into manageable, well-scoped sessions. By enforcing a 'one session, one spec' philosophy, it limits work to 2-4 hour blocks with clear task checklists, ensuring projects stay within AI context windows while maintaining rigorous standards through automated state tracking, PRD synchronization, and mandatory validation stages. It is ideal for teams and solo developers who need to maintain institutional memory and project architecture while working with LLMs.

Key Features

0117 GitHub stars

02Structured 13-command workflow covering initialization, implementation, and phase transitions.

03Deterministic state tracking via a centralized .spec_system directory and state.json file.

04Context-optimized session scoping limited to a maximum of 4 hours and 25 tasks per session.

05Strict ASCII encoding enforcement to ensure maximum compatibility with AI model tokenization.

06Integrated monorepo support with package-aware session mapping and cross-cutting coordination.

Use Cases

01Standardizing AI development workflows across a team to ensure consistent documentation and quality.

02Maintaining a live, accurate Product Requirements Document (PRD) that synchronizes with codebase progress.

03Scaling complex feature builds by decomposing them into manageable, trackable implementation sessions.

Key Features

0117 GitHub stars

02Structured 13-command workflow covering initialization, implementation, and phase transitions.

03Deterministic state tracking via a centralized .spec_system directory and state.json file.

04Context-optimized session scoping limited to a maximum of 4 hours and 25 tasks per session.

05Strict ASCII encoding enforcement to ensure maximum compatibility with AI model tokenization.

06Integrated monorepo support with package-aware session mapping and cross-cutting coordination.

Use Cases

01Standardizing AI development workflows across a team to ensure consistent documentation and quality.

02Maintaining a live, accurate Product Requirements Document (PRD) that synchronizes with codebase progress.

03Scaling complex feature builds by decomposing them into manageable, trackable implementation sessions.