with one click
Spec-to-Ship
Spec-to-Ship contains 12 collected skills from mrtblount, with repository-level occupation coverage and site-owned skill detail pages.
Skills in this repository
Spec-Driven Development protocol that runs before every new project or major feature. Creates structured, LLM-optimized specifications that enable one-shot implementation. Guides through 4 sequential phases: Specify, Plan, Tasks, Implement. Detects when specs are needed vs. when to just build. After spec creation, identifies and installs required tools and MCPs. Use when: "new project", "build me an app", "I want to create", "spec this out", "let's spec", "start a new project", "build a spec", "one-shot this", "new feature", "add a feature", "build a feature", "I want to build", "I want to add", "major feature", "big feature", "implement a feature", "add X to the project". MUST trigger proactively when Claude detects a request that would touch 5+ files, introduce new data models or APIs, or require significant new functionality — even if the user does not explicitly mention specs or SDD. Do NOT start implementing multi-file features without running this skill first. If in doubt about scope, run Phase 0 (Scope
Phase 2 of Spec-Driven Development. Translates approved SPEC.md into a comprehensive technical plan covering tech stack, data models, API contracts, file structure, and architectural constraints. Output: PLAN.md optimized for LLM implementation.
Phase 3 of Spec-Driven Development. Transforms approved SPEC.md and PLAN.md into atomic, dependency-ordered tasks. Each task touches 1-3 files maximum and is independently testable. Output: TASKS.md optimized for sequential LLM execution.
Deploys Spec-Driven Development into an existing project. Scans codebase, build logs, CLAUDE.md, git history, and file structure to reconstruct a baseline understanding of the project's current state. Produces BASELINE.md that anchors all future spec work. Use when applying SDD to projects already in progress.
Phase 1 of Spec-Driven Development. Interactive specification gathering that produces a comprehensive SPEC.md covering business logic, user journeys, edge cases, success criteria, and invariants. Optimized for LLM consumption.
Syncs SDD outputs to project management tools. Pushes TASKS.md to GitHub Projects as issues, SPEC.md to Linear as documents/initiatives. Handles bidirectional sync setup between GitHub Projects and Linear. Use after specs are finalized.
Post-spec tool discovery, environment setup, and pre-flight checklist agent. Analyzes SPEC.md, PLAN.md, and TASKS.md to identify all required tools, MCPs, packages, and services. Inventories ALL available tools in the session. Automatically installs what it can, then produces a pre-flight checklist so the user can confirm everything is ready BEFORE autonomous implementation begins.
AI-native Product Requirements Document builder for new and existing products. Walks through structured discovery, market research, and generation phases to produce comprehensive PRDs aligned with the 7-section accelerator framework. Detects existing codebases and pulls context before interviewing the user. Use when: "create a PRD", "product requirements", "write a PRD", "PRD for", "new product", "product spec", "requirements document", "define the product", "what should we build", "product definition", "I need a PRD". Triggers on: PRD, product requirements document, product definition, product spec, define requirements, product planning.
Phase 1 of PRD Builder. Scans existing codebases to understand what's built, then walks the user through a structured discovery interview covering problem, target user, features, priorities, and AI considerations.
Phase 3 of PRD Builder. Synthesizes discovery notes and research into a complete Product Requirements Document using the 7-section accelerator framework with AI-native extensions.
Phase 2 of PRD Builder. Conducts web research to gather market context, competitive landscape, user patterns, and relevant insights that strengthen the PRD with real-world grounding.
Phase 4 of PRD Builder. Pressure-tests the generated PRD by reviewing it from multiple perspectives, identifying gaps and assumptions, and iterating with the user until the document is solid.