Back to directory

Agent Skills for WordPress

A set of portable Agent Skills that teach AI coding assistants such as Claude, Copilot, Codex and Cursor expert-level WordPress development: Gutenberg blocks, block themes, plugin architecture, the REST and Abilities APIs, WP-CLI and performance profiling.

Agent Skills for WordPress

Agent Skills for WordPress is a collection of instructions, checklists and scripts that give an AI coding assistant expert-level knowledge of WordPress development conventions, aiming to stop assistants from generating outdated pre-block-theme patterns, missing security considerations, or skipping proper block deprecations.

Individual skills cover classifying a WordPress repository and routing to the right workflow, Gutenberg block and block theme development, plugin architecture and the settings API, the REST API and the newer Abilities API for capability-based permissions with matching audit and verification skills, WP-CLI automation and multisite operations, performance profiling and caching, PHPStan static analysis configured for WordPress typing, and using WordPress Playground.

The skills were generated from official Gutenberg and WordPress documentation using an AI coding model, then reviewed and edited by WordPress contributors, and are published with an explicit AI-authorship disclosure. It is written mainly in JavaScript.

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.