Spec-Kit

Source Excerpt

Executive Summary

Spec Kit is GitHub's open-source toolkit for spec-driven development — a methodology that flips traditional software development by making specifications executable and directly generating working implementations. Instead of specs being disposable scaffolding, they become the source of truth that drives code generation through AI agents.

Strategic Value for retired internal project: Directly validates and operationalizes the Superpowers Method that Julius follows. Provides official GitHub-backed tooling for the spec-first approach we already practice. Critical for standardizing agent workflows across the Quad.

Core Philosophy

What is Spec-Driven Development?

Traditional development: Code is king → Specifications are scaffolding discarded after implementation.

Spec-driven development: Specifications are executable → Specs directly generate working implementations.

The Flip

"Build high-quality software faster by focusing on product scenarios and predictable outcomes instead of vibe coding every piece from scratch."

Installation & Setup

Specify CLI (Main Tool)

Persistent Installation (Recommended):

# Install latest stable release
uv tool install specify-cli --from git+https://GitHub.com/GitHub/spec-kit.git@vX.Y.Z

# Or install from main (unreleased changes)
uv tool install specify-cli --from git+https://GitHub.com/GitHub/spec-kit.git

One-time Usage:

uvx --from git+https://GitHub.com/GitHub/spec-kit.git@vX.Y.Z specify init <PROJECT_NAME>

Workflow Integration with AI Agents

Spec Kit exposes slash commands for AI agents:

The Spec-Driven Workflow

Phase 1: Establish Principles

Command: /speckit.constitution

Create governing principles and development guidelines:

/speckit.constitution Create principles focused on code quality, testing standards,
user experience consistency, and performance requirements

Output: CONSTITUTION.md — Project's foundational constraints and values.

Phase 2: Create Specification

Command: /speckit.specify

Describe what and why, not tech stack:

/speckit.specify Build an application that can help me organize my photos in separate
photo albums. Albums are grouped by date and can be re-organized by dragging and dropping.
Albums are never nested. Photos previewed in tile interface.

Output: SPEC.md — Functional specification with user scenarios.

Phase 3: Technical Implementation Plan

Command: /speckit.plan

Provide tech stack and architecture choices:

/speckit.plan The application uses Vite with minimal libraries. Use vanilla HTML, CSS,
and JavaScript. Images not uploaded; metadata stored in local SQLite database.

Output: PLAN.md — Technical architecture and implementation strategy.

Phase 4: Task Breakdown

Command: /speckit.tasks

Generate actionable task list from plan:

/speckit.tasks

Output: TASKS.md — Ordered, granular implementation tasks.

Phase 5: Execute Implementation

Command: /speckit.implement

Execute all tasks and build feature according to plan:

/speckit.implement

Output: Working implementation matching spec.

Community Extensions

Spec Kit supports extensible workflows via community extensions:

Extension Categories:

• docs — Reads, validates, or generates spec artifacts
• code — Reviews, validates, or modifies source code
• process — Orchestrates workflow across phases
• integration — Syncs with external platforms (Jira, Confluence, Azure DevOps)

Supported AI Agents

Spec Kit works with multiple AI coding agents:

Development Phases (Roadmap)

Spec Kit itself is developed in phases:

Alignment with retired internal project

Superpowers Method Validation ✅

Spec Kit validates and operationalizes the Superpowers Method that Julius follows:

Integration Opportunities for retired internal project

1. Hermes Skill Development (High Priority)

• Create /speckit slash commands for Hermes Agent
• Map Superpowers Method phases to Spec Kit workflow
• Enable Julius to delegate spec-driven tasks with standardized structure

2. Quad Workflow Standardization

• Adopt Spec Kit as standard for all agent-generated code
• Cypher: Use for platform feature development
• Reacher: Use for test/benchmark specification
• Octavius: Use for skill documentation and learning artifacts

3. Extension Development (Medium Priority)

• Build retired internal project-specific extensions:
• retired-internal-project-governance — Julius approval gates before implementation
• retired-internal-project-validation — Reacher benchmark integration
• retired-internal-project-memory — Auto-archive to wiki on completion

Key Insights

1. GitHub Backing = Legitimacy

Spec Kit isn't a side project — it's official GitHub Open Source with dedicated team, roadmap, and enterprise support. This validates spec-driven development as an industry pattern.

2. Extension System = Ecosystem Play

The community extensions catalog (20+ extensions already) shows Spec Kit is designed as a platform, not just a tool. Extensions cover:

• Process orchestration (AIDE, Fleet, Conduct)
• Quality gates (Cleanup, DocGuard, FixIt)
• Integration (Jira, Confluence, Azure DevOps)
• Learning & documentation (Learning Extension, Archive)

3. Agent-Agnostic Design

Spec Kit works with Claude Code, Codex, OpenCode, and others via slash commands. This agent-agnostic approach aligns with retired internal project's Multi-Agent Orchestration architecture.

4. Human-in-the-Loop Gates

Extensions like Fleet Orchestrator explicitly support human review gates — critical for governance-heavy environments like retired internal project.

5. Enterprise Ready

Air-gapped installation guide, OAuth integrations (Azure DevOps), and audit-friendly workflows show Spec Kit is built for production enterprise use, not just hobby projects.

Source excerpt truncated at 220 of 292 lines. Open the canonical wiki path above for the full page.