Back to directory

Code Addiction (ADD)

A spec-driven development framework of commands, scripts and skills for AI coding assistants that turns a feature idea into discovery, planning, review and delivery through a fixed sequence of slash commands.

Code Addiction (ADD)

Code Addiction standardizes AI-assisted development, which is otherwise fragmented across custom prompts and editor-specific conventions, into a shared core of commands, scripts, skills and templates, with provider-specific integrations for Claude, Codex, Antigravity, Cursor and OpenCode, installed and managed by a versioned CLI with install, update, uninstall, doctor and validate commands.

Every feature follows the same development trail from idea to delivery; each stage's command hands off to the next, and steps can run unattended by approving a brainstorm stage with a delivery flag, with a final review and pull-request prompt always waiting for explicit developer approval. Optional plugins integrate external MCP tools to extend discovery, planning and diagnosis, disabled by default until the corresponding external tool is installed.

It is written in JavaScript and MIT licensed, and works with NestJS, React or other stacks and databases without requiring a specific tech stack.

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.