Spec Kit

Active
GitHub Python MIT

Description

GitHub's official open-source Spec-Driven Development toolkit that flips traditional software development — specifications become executable artifacts that directly generate working implementations, integrating with Copilot, Claude Code, Cursor, and other AI coding agents.

Key Features

  • Spec-Driven Development — specifications become executable artifacts that directly generate implementations
  • Specify CLI — initialize projects and establish principles with a single command
  • Multi-agent integrations — supports GitHub Copilot, Claude Code, Cursor, and other AI coding agents
  • Self-management upgrade — built-in self check / self upgrade commands to manage CLI versions
  • Template-driven workflow — define planning, implementation, and validation phases via YAML workflows
  • Extensible presets — customize Spec Kit with team-specific extensions and presets

Use Cases

💡 Making AI coding agents follow specifications rather than generate code freely, improving predictability
💡 Standardizing the plan -> implement -> validate workflow across teams
💡 Replacing vibe coding with structured requirement management
💡 Scaling software delivery with GitHub Copilot, Claude Code, and other agents
💡 Providing consistent specs and templates during new project initialization

Strengths & Limitations

Strengths

  • Actively maintained, recent updates
  • High community interest (133.1k stars)
  • Permissive open-source license (MIT)
  • Responsive to issues, low backlog

Quick Start

# Requires uv (https://docs.astral.sh/uv/)
uv tool install specify-cli --from git+https://github.com/github/spec-kit.git@v0.0.20
specify init my-project --integration copilot
cd my-project
specify self upgrade  # keep the CLI in sync with the latest release

Related Projects

AI-DLC Workflows

4.3k · TypeScript
Active A

AWS Labs workflow steering rules for AI-Driven Life Cycle development, guiding AI coding agents through adaptive software-delivery processes.

coding-agentworkflowaws +1
  • · Three-phase adaptive workflow (Plan → Implement → Validate) for AI-driven development
  • · Multi-platform support: Kiro, Amazon Q, Cursor, Cline, Claude Code, GitHub Copilot
  • · Steering rules that guide AI coding agents through structured software delivery

Claude Code Workflow

2.1k · TypeScript
Normal B

JSON-driven multi-agent cadence-team development framework with intelligent CLI orchestration (Gemini/Qwen/Codex), context-first architecture, and automated workflow execution.

claude-codeworkflowmulti-agent +1
  • · JSON-driven multi-agent framework with skill-based workflows from lite-plan to brainstorm modes
  • · Multi-CLI orchestration supporting Gemini, Codex, Qwen, OpenCode, and Claude with semantic invocation
  • · Team Architecture v2 with role-based agents and inner loop execution patterns

n8n Skills for Claude Code

6.2k · Shell
Active A+

A skillset for Claude Code to build flawless n8n workflows, letting AI assistants directly generate usable n8n automation workflows.

n8nclaude-codeskills +2
  • · Claude Code integration - directly attach as skill package to Claude Code
  • · Full n8n node coverage - built-in schema and usage examples for hundreds of n8n nodes
  • · Workflow generation - input business requirements, auto-generate complete n8n JSON workflows

Helmor

1.3k · TypeScript
Active B

Helmor is a local-first multi-agent coding workbench that runs Claude Code, Codex, and other agents in parallel across isolated git worktrees.

workflowmulti-agentcoding-agent +2
  • · Isolated git workspaces - every task gets its own worktree and branch so main stays clean
  • · Multi-agent runtime - Claude Code, OpenAI Codex, Cursor, OpenCode run side by side
  • · Mid-flight steering - watch live output, inject follow-ups, queue prompts, mention files with @