github/spec-kit

💫 Toolkit to help you get started with SDD or any other process!

What it solves

Spec Kit provides a structured framework for AI coding agents to follow, preventing them from jumping straight into code without a plan. It solves the problem of inconsistent or unplanned AI-generated software by enforcing rigorous processes for feature development, bug fixing, and idea validation.

How it works

It integrates with AI coding agents (like GitHub Copilot) as a set of "skills" (slash commands). Instead of free-form chatting, users guide the agent through predefined workflows:

  • Spec-Driven Development (SDD): A pipeline that moves from a project constitution to specification, technical planning, task breakdown, implementation, and final convergence.
  • Bug Fixing: A separate three-step process (assess $\rightarrow$ fix $\rightarrow$ test) to ensure the cause is diagnosed before the fix is applied and verified.
  • Idea Assessment: A research-heavy workflow (intake $\rightarrow$ research $\rightarrow$ define $\rightarrow$ shape $\rightarrow$ decide) to determine if a feature is worth building based on evidence.

Who it’s for

Developers using AI coding agents who want more predictable, high-quality outcomes and documented technical decisions.

Highlights

  • Structured Workflows: Replaces haphazard prompting with repeatable processes for building, fixing, and assessing.
  • Agent Integrations: Works with multiple AI coding agents via integration keys.
  • Extensible: Supports custom extensions, presets, and bundles to adapt the process to specific team needs.
  • Documented Outcomes: Generates Markdown artifacts in the project directory (e.g., .specify/) to track the evolution of a feature or bug fix.

Related

  • Project
    spec-kitty/spec-kitty

    A CLI for spec-driven development that turns product intent into a governed AI coding workflow with repo-native tracking and git worktree isolation.

  • Project
    kaochenlong/spectra-app

    A Spec-Driven Development tool consisting of a desktop app, CLI, and AI agent skills to visually track and manage specifications, change proposals, and tasks.

  • Project
    MageByte-Zero/spec-superflow

    A lightweight, recoverable AI programming workflow tool that enforces structured planning and formal verification to ensure AI-generated code changes are correct and within scope.

  • Dispatch
    Solving AI Drift with Spec-Driven Development (SDD)

    Explore how a new Claude skill implements a shared source of truth to prevent contradictory behavior across multiple AI coding tools like Cursor, Claude Code, and Copilot.

  • Project
    Lamatic/AgentKit

    AgentKit is an open‑source SDK that provides a catalog of 70 ready‑made “kits” (full apps, pipelines, or templates) for building, testing and deploying AI agents. Kits are defined by a simple config file, can be edited in a visual flow studio, and are shipped as Next.js apps that run server‑less. The collection covers support triage, code review, RAG chat, hiring assistants, legal/medical bots, and many other business‑focused agents, making it a turnkey platform for creating reliable, enterprise‑grade AI agents.