# Spec-kit: The Open Toolkit

> A hands-on look at GitHub's Spec Kit  how it structures the Specify → Plan → Tasks workflow and integrates with PR review gates.

- **Source:** https://ainativesoftware.engineering/roadmap/day-5/spec-kit-the-open-toolkit
- **Site:** AI-Native Software Engineering — https://ainativesoftware.engineering/book

- **Day:** 5 · position 3 of 4
- **Reading time:** 1 minute
- **Day overview:** [Day 5](https://ainativesoftware.engineering/roadmap/day-5.md)

A hands-on look at GitHub's Spec Kit  how it structures the Specify → Plan → Tasks workflow and integrates with PR review gates.

## Theory

**Spec-kit** is an open-source toolkit from GitHub that operationalizes Spec-Driven Development inside a standard GitHub workflow.


### What Spec-kit provides

- A structured template set: spec files, plan files, and task checklists
- PR workflow integration  specs live alongside code and are reviewed like code
- Review gates that keep product intent aligned with what's actually shipped
- A lightweight, file-based approach that works in any repo without additional tooling


### The Spec-kit workflow

1. **Specify** Write a spec file describing intent, constraints, and acceptance criteria
2. **Plan** Generate or author a plan file with the implementation approach
3. **Tasks** Break the plan into atomic, checkable tasks
4. **PR** Open a pull request; the spec is the contract reviewers check against


### When Spec-kit shines

Spec-kit works well for teams that want a low-ceremony, Git-native SDD approach without adopting a full framework. It integrates naturally into existing GitHub workflows and is a good starting point before evaluating heavier frameworks like BMAD.

**Theory resources**

- [Spec Kit - Official Toolkit](https://speckit.org/)
- [Getting Started with Spec-Kit - Microsoft Developer Blog](https://developer.microsoft.com/blog/spec-driven-development-spec-kit)

## Practice

Run this in a repository you already know, not a toy project.

1. Install Spec-kit in a test repository following the official getting-started guide.
2. Pick a small, real feature (a new API endpoint, a UI component, or a CLI command).
3. Use Spec-kit's templates to write a spec, generate a plan, and break it into tasks.
4. Hand the tasks file to an AI agent and have it implement the first task.
5. Open a draft PR and verify the spec is visible in the PR description.

- **Previous topic:** [SDD as a Pillar of AI-Native Engineering](https://ainativesoftware.engineering/roadmap/day-5/sdd-as-a-pillar-of-ai-native-engineering.md)
- **Next topic:** [Spec-kit Under the Hood](https://ainativesoftware.engineering/roadmap/day-5/spec-kit-under-the-hood.md)

---

_AI-Native Software Engineering by Alfonso Graziano (O'Reilly Media, Early Release; print edition February 2027). Every page of ainativesoftware.engineering is also served as Markdown: append `.md` to any URL. Index: https://ainativesoftware.engineering/llms.txt — whole site in one file: https://ainativesoftware.engineering/llms-full.txt._