Back to directory

Spec Kit

A toolkit that gives AI coding agents structured, reusable processes for building features from a specification, fixing bugs or assessing an idea.

Spec Kit

Spec Kit is an open-source toolkit that gives AI coding agents structured processes, reusable templates and documented outcomes instead of open-ended prompting. It offers three independent processes a user can start from: Spec-Driven Development, which carries a specification through planning, implementation and convergence; bug fixing, which produces an assessed cause, a scoped fix and a recorded verification; and idea assessment, which produces an evidence-backed decision to go ahead, clarify or stop.

Spec-Driven Development ships as the toolkit's core, while bug fixing and idea assessment are separate installable extensions. Each process is invoked as a set of agent skills, run one at a time in the user's coding agent, with the user reviewing the result before continuing. The project is installed with a Python CLI (specify-cli) and supports several coding agents, including GitHub Copilot.

More in Agent skills

See all 581
  • Provenance gate

    github.com

    A deterministic authorization gateway that sits between an AI agent and its tools, tracing where each parameter of a side-effect call came from, and blocking the call or asking a human when the source does not check out.

  • A markdown framework of system rules that restates Isaac Asimov's Four Laws of Robotics as a set of prioritized loss calculations for large-language-model agents to follow, published as a discovery file for agents and crawlers to read.

  • Vitruvius

    github.com

    An engineering research agent that runs a discover, read, synthesize, verify and review loop over technical questions, reading sources directly rather than inferring from titles, and publishing its own scored verifier benchmark, weaknesses included.