GitHub Spec Kit
GitHub's open-source toolkit for spec-driven development, where a written specification drives AI coding agents.
Overview
With Spec Kit, an open-source (MIT) toolkit from GitHub, you stop prompting agents directly. Instead, you write a specification, plan, and tasks that serve as a source of truth for agents to build, test, and validate against. Announced September 2025, it has grown to over 120K GitHub stars and integrates with more than 25 coding agents including Claude Code, Codex CLI, Copilot, Cursor, and Gemini CLI. Its workflow runs through commands like specify, plan, tasks, and implement.
Pricing
Pricing shown for reference only. These figures reflect RECATOOLS research as of 24 Jul 2026 and may be out of date or incomplete. This is not financial or purchasing advice — always confirm the current price on the provider’s official website before making any decision.
ASEAN Perspective
GitHub Spec Kit in Southeast Asia
ASEAN-region availability and pricing notes coming soon. Drop the editorial team a note via /contact/ if you can supply local context (Singapore/Malaysia/Indonesia/Thailand/Vietnam).
What this is for: Structuring AI-assisted development around a written spec so agents build to an explicit contract rather than an ad-hoc prompt.
Who this is for: Teams and developers who want repeatable, reviewable AI coding instead of unstructured 'vibe coding'.
Availability: Free and open-source (MIT) on GitHub; works with 25+ coding agents.
What people say
Spec Kit rode a huge wave of interest after its September 2025 launch, cracking into GitHub's most-starred repos and getting picked up by Microsoft Learn as the reference implementation of "spec-driven development." Practitioners who like structure over pure "vibe coding" describe it as a sensible middle ground: keep AI speed but force a specify, plan, tasks, implement contract that agents build against.
The criticism is pointed. A well-circulated GitHub Discussion argues SpecKit "creates the illusion of work, generating a bunch of text," with people reporting hours spent correcting LLM-authored specs because the model makes many mistakes when forming them. On multi-module brownfield systems, reviewers say the workflow produces volume rather than fidelity. There's also framework churn: the v0.10.0 release removed the entire --ai flag family, breaking tutorials and scripts written only months earlier.
It's a promising discipline for greenfield or well-scoped work and a good fit for teams that want reviewable AI output. The overhead is non-trivial, however, and its payoff on large legacy codebases is disputed.
Summary of public user & expert reviews, compiled by RECATOOLS.
About this listing
This entry was compiled from publicly available data including GitHub Spec Kit's official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with GitHub Spec Kit unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to GitHub Spec Kit directly →
Spotted something out of date? Suggest an update →
Alternatives to GitHub Spec Kit
More in Code & Dev Tools
