About this project

Spec Kit is an open-source toolkit from GitHub that helps teams use Spec-Driven Development (SDD) with AI coding agents. Instead of treating specifications as throwaway scaffolding, SDD makes specifications executable: they directly generate working implementations. The toolkit provides a ready-to-use spec-driven process that can be customized or replaced with your own workflow. The main component is the Specify CLI, installed via uv or PyPI. After running `specify init`, it sets up a project and installs slash commands or agent skills for common AI coding agents. Supported integrations include 30+ CLI and IDE-based coding agents, such as Copilot, Codex, Claude Code, and others. Core slash commands follow a structured workflow: - `/speckit.constitution` — establish project principles and development guidelines - `/speckit.specify` — define what to build (requirements and user stories) - `/speckit.plan` — create a technical implementation plan - `/speckit.tasks` — break the plan into actionable tasks - `/speckit.taskstoissues` — convert tasks into GitHub issues - `/speckit.implement` — execute tasks to build the feature - `/speckit.converge` — compare the implementation against the spec, plan, and tasks, and append remaining work as new tasks Optional commands include `/speckit.clarify` for clarifying underspecified areas, `/speckit.analyze` for cross-artifact consistency checks, and `/speckit.checklist` for custom quality checklists. The repo also bundles opt-in extensions. A bug-fixing extension adds an assess → fix → test workflow for scoped, evidence-based bug fixes. An idea-assessment extension provides an intake → research → define → shape → decide workflow to turn raw ideas into documented go / needs-clarification / kill decisions. Customization is supported through three mechanisms: extensions add new commands and templates, presets override how existing templates and commands behave, and bundles package curated sets of extensions, presets, steps, and workflows into role-based setups (e.g., product manager, business analyst, security researcher, developer). Project-local overrides allow one-off adjustments. Templates are resolved at runtime, while extension and preset commands are applied at install time. Spec Kit also includes self-management commands such as `specify self check`, `specify self upgrade --dry-run`, and `specify self upgrade`, with options to pin specific release tags. Community resources include extensions, presets, bundles, walkthroughs, and related projects on the docs site. The project dogfoods itself: the maintainers use Spec Kit when developing Spec Kit, particularly for substantial features, via an automated feature assessment workflow.